BA_Algorithms by Bhupinder Anwar https://padlet.com/banwar1/cmz5k27huynx9kdf Post an example of Algorithm used in the real world. Include an image and short description en-us 2021-02-24 02:18:08 UTC 2024-06-02 20:25:52 UTC hello@padlet.com https://padlet.net/icons/png/1f4f1.png College Admissions Algorithms banwar1 https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1234426776 Companies are selling admission algorithms arguing they are fair. Analysis of student's  data and  college's definition of 'student success' will drive the analysis.  The  companies are looking at even your game playing behavior to 'tease out' and measure chosen traits. Basically anything a student is doing online - posting, watching videos , gaming,  posting student centered videos can be used to measure desired  student characteristics.  Selected data  can be used to make predictions about student  and their  chance of accepting the admission to that college. 

Negative consequences of these algorithm include data bias and data privacy issues. I will add more here

For more information: 
https://www.edsurge.com/news/2020-07-10-as-colleges-move-away-from-the-sat-will-admissions-algorithms-step-in]]> 2021-02-24 02:27:00 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1234426776 Boeing 777 Integrated Checklist mvbmccumiskey https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237748833 The Boeing 777 checklists can calculate the engine temperatures and can detect if there is an engine failure/fire. If there is a certain high temperature, it's an indicated Engine failure. That then prompts a full checklist for the pilots to safely take the aircraft to the ground, and contain the engine fire. 

]]> 2021-02-24 18:14:13 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237748833 AI Detecting Students’ Emotions While Learning mvapatra https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237805415 A startup called Find Solution AI in Hong Kong launched 4 Little Trees, a software designed to detect a student’s emotions to help with their learning. While using the software, the AI uses an algorithm that detects the student’s emotion by measuring muscle points. The software also takes their results on assignments and creates a report on their strengths and weaknesses. With this data, tests designed like a game are offered for the students to make school fun. Founder Viola Lam says students using this software perform 10% better than if they didn’t use the program.

There are positive benefits, as exam grades can be improved through this software. With the reports on the student’s strengths and weaknesses and data about the student’s emotions, this can definitely help the student’s learning process. The program can detect primary emotions, like happiness, sadness, or disgust, with a high accuracy. However, this software cannot detect secondary emotions, such as discomfort or signs of anxiety. Privacy concerns also arise, as students may not feel comfortable showing their face while using the software. Overall, this program can be useful for students to improve exam scores, especially with online learning in the current pandemic.

Article Link: https://www.cnn.com/2021/02/16/tech/emotion-recognition-ai-education-spc-intl-hnk/index.html 

Video Link(linked below): https://www.cnn.com/videos/business/2021/02/15/emotion-recognition-students-find-solution-ai-spc-intl-hnk.cnn/video/playlists/business-evolved/ 
]]> 2021-02-24 18:25:09 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237805415 Pacman Ghost Algorithm mvijain https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237809830 In Pacman the ghosts that run towards the pacman are controlled by an algorithm using x and y coordinates. Each ghost has a different sort of "personality" which means they have different algorithms that decide how and where they move. The ghosts are driven by algorithms about certain tiles that they have to reach in the maze.

The impact of this algorithm is that the ghosts all have different ways that they move around the maze in pacman, and if the algorithms are studied it is possible to find patterns in their movements to understand how to beat the game faster and easier.

https://gameinternals.com/understanding-pac-man-ghost-behavior
101computing.net/pacman-ghost-algorithm/

]]> 2021-02-24 18:25:57 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237809830 Algorithms in Sports mvekuo https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237829095 Sports teams in leagues such as the NFL, MLB, and NBA have recently been using algorithms and advanced statistics to determine the best coaches to hire, players to sign, and in basketball what shots to take. Nowadays massive amount of athlete data are collected through things like cameras and wearable devices that create refined performance metrics that gives coaches all the information they need on players such as when to rest them, what lineups are the most effective, and who needs to score more for the team to win.]]> 2021-02-24 18:29:34 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237829095 Netflix Recommendation Algorithms mvkau https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237829552 There are thousands of movies and shows on Netflix which makes it hard for a user to find one that suits them. To assist that, Netflix uses an algorithm that is based on information about media you've watched (genre, actors, categories etc.), when you watch, how long you watch and some other factors to create a recommendation list personalized for you. By using this information, the algorithm filters out different shows and movies that fit your description. This is a great algorithm that all streaming services need because finding everything on your own with no help will be tiring and difficult.

For more information: https://help.netflix.com/en/node/100639]]> 2021-02-24 18:29:39 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237829552 Tinder Swiping Algorithm mvmrubendall https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237916734 To help people find good matches,  Tinder has an algorithm. The algorithm primarily focuses on location and age; you get matched with your preferences.  Tinder has something called "super likes" that enable you to like someone "louder" aka when they see your profile back, there will be a blue star indicating that they swiped right on you. The algorithm then has to pause and add your profile with the super like closer to the top of the pile of profiles the other person is given. As you swipe through profiles and get closer to the end, the Tinder algorithm starts to recycle previous people, as someone you rejected before might seem better after looking through thousands of other people. It will also recycle previous match ups, too. Although love isn't something you can solve with an algorithm, the Tinder algorithm does help some people find others.]]> 2021-02-24 18:45:50 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237916734 Risk Assessment Algorithm, mvsbrandeis2 https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237949289 Risk Assessment algorithms are formulas that predict a convict's likelihood of recidivism. The algorithms are very prevalent in the criminal justice system and can be used to replace cash bail systems, determine sentence length, and even decide conditions of parole and probation. These risk assessments use factors like age, criminal history, and employment to make judgements on whether a criminal will commit another crime. This algorithm could be incredibly useful in replacing the outdated and regressive bail system, allowing those of lower socioeconomic classes who are lower risk to  spend time at home instead of in jail before their trial. 

However, in practice, this algorithm has proven to be inaccurate and racially biased. A study done by the Equal Justice Initiative found that just 20 percent of those who were chosen as likely reoffenders were actually charged with another crime. Furthermore, Black defendants were 77% more likely to be chosen as high risk, and 45% more likely than white defendants to be predicted as a recidivist.  ]]> 2021-02-24 18:52:02 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1237949289 Online Shopping Algorithm mv_alee2 https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1238813112 Have you ever been trying to buy something online, and then something pops up on the side that catches you attention?  This is due to product recommendation engines that use algorithms to entice their buyers with certain products.  These algorithms take search history, shopping history, and other customer's purchase history into account, and then filter through their products to give you suggestions based on what they think you would like.  

These algorithms are extremely beneficial to retailers because they cause customers to spend more money.  Amazon stated that 35% of their sales were due to the recommendation engines.

https://www.theukdomain.uk/online-retailers-can-use-algorithms-grow-business/]]> 2021-02-24 22:58:36 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1238813112 New AI algorithm that automates artificial pancreas system mvjjones https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1238928528 A research team from POSTECH's Department of Convergence IT Engineering and Electrical Engineering has created an AI based algorithm for an artificial pancreas system. The pancreas' are used by people with type 1 diabetes, where they have to input every meal intake of carbohydrates, so the pancreas can produce insulin. The algorithm (AlphaGo) makes this much simpler and easier as it automatically gives the person the required amount of insulin that they need without needing to input the intake.

The algorithm works by eliminating the inconvenience of inputting the meal intake by sensing the glucose level of the person, then releasing how much insulin the person needs to deal with it. The algorithm has the beneficial effect of  "enabl[ing] fully automated blood sugar control without the hassle of inputting meal or exercise information." This saves a lot of time for a person and could potentially spread to other treatments, making them more efficient as well.

https://www.news-medical.net/news/20210223/New-AI-algorithm-to-fully-automate-the-artificial-pancreas-system.aspx]]> 2021-02-24 23:59:52 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1238928528 Self Driving Car Algorithm mvmsharp https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1239432923 Every year self driving cars seem to become more accessible and popular so it is important for them to have very good algorithms because of how much they have to manage.  The algorithm doesn't just hold the data and commands that it is given but it actually learns new things as it is used. This means that the algorithm can notice and learn about little things that are important while driving such as navigation, obstacles that may interfere with the car, and even the drivers behavior based on their driving.  This ability for the algorithm to learn is known as machine learning and the way it works is done through a process called supervised or unsupervised learning. These types of learnings use a combination of predictions, input data, output data, clustering, classification, and regression to allow the car to learn.

https://medium.com/@intellias/how-machine-learning-algorithms-make-self-driving-cars-a-reality-4180a528121c]]> 2021-02-25 04:53:44 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1239432923 Spotify Song Recommendation Algorithm mvlbennett https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1239555560 Millions of people around the world listen to music. One of the most popular applications that people use to listen to music is Spotify. There is a good reason for this too.  A big reason why Spotify is so popular is that it is unique to each user. To make the listener's experience more enjoyable, Spotify recommends songs, playlists, artists, and albums to the user using BaRT.  BaRT takes into account things like what you listen to, your music history, your age, location, and the artists or genres of music that you listen to the most. If you listen to any of the songs or playlists that it offers you for more than 30 seconds, the algorithm sees that as a successful recommendation and will continue to recommend music like that to you. BaRT also keeps diversity within the recommended playlists. It does this by taking into account what other people your age, similar location, and people around the world are currently listening to. It then offers the music to you under a title like "Something New". I think that the impact of this can only be positive because it makes the experience much more enjoyable, unique, and special for the user.

https://onezero.medium.com/how-spotifys-algorithm-knows-exactly-what-you-want-to-listen-to-4b6991462c5c


]]> 2021-02-25 06:02:54 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1239555560 Youtube Recommendation Algorithm mvajiang3 https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1242732182 The platform Youtube pulls billions of users per month. Its recommendation algorithm uses the users watch history to further recommend videos for the user. It is this algorithm that makes Youtube so addicting and it is reported that the algorithm is what decides what people watch 70% of the time. Youtube says that its algorithm is a "real time feedback loop that tailors videos to each viewer's different interests." To use the recommendation algorithm, Youtube first ranks videos based on performance analytics data. This data contains clickthrough rate, watch through rate, retention rate, engagement (likes, dislikes, comments), view rate of growth, and more factors. It then matches these videos to people. This matching system is based on the users watch history and what people of similar interests have watched. For this, they look at data from the user pertaining to what channels and topics they watch, how much time they spend watching topics, how many times the video has appeared already, and what they don't watch. The user activity not only influences the users recommendation stream, but also what pops up for the users search results. As a result of the recommendation algorithm, Youtube's platform has gained a lot of traction and users since its release.
The image below is simple flowchart released by Youtube in 2016 regarding its recommendation algorithm:
Source: https://blog.hootsuite.com/how-the-youtube-algorithm-works/]]> 2021-02-25 18:58:54 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1242732182 TikTok Recommendation Algorithm mvayashiki https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243434480 The social media app TikTok, uses an algorithm for the users ForYou page. The videos that come up on your ForYou page is specialized to each user. The factors that are included in choosing the next video for you to watch are based off your interactions, the videos information(content), and your device and account settings. The TikTok algorithm pays attention to every little thing, like how long you watch the video from begging to end, so everything is in account for. Also, TikTok algorithm is used to put your videos on the ForYou page. It is more likely to occur if your a brand, business, or a well known creator already on the app. Here are a few factors if you want your video on the ForYou page that the TikTok algorithm uses, if you use more hashtags in your video, the most popular ones at the time will work better(look on the discover tab). Another factor is what your caption is, ones that are like "Part 1", or one that makes it seem more mysterious. Lastly, the TikTok algorithm will more likely pick your video if you use the trending sounds or songs. As TikTok uses this algorithm, it it used to keep you hooked on the app so you spend more time on it.
This image below shows types of content that are on peoples ForYou page.
Source:https://later.com/blog/tiktok-algorithm/ 

]]> 2021-02-25 21:55:12 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243434480 Page Rank Algorithm mvjmckinley https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243610644 Page Rank was a program that Google used to decide what websites showed up first when something was searched. The way the algorithm would work is it would assign a website a number ranging from 0 to 10 based on how many links were pointed to the website. The higher the number the more important and authoritative the website was deemed by the algorithm. Resulting in the website being ranked higher in search results. The more links that pointed to a website the higher the number that was assigned became. 0 is a low-quality website and 10 being a very high-quality website. At the time this was not the only program that was used to determine the rank of search results. But it has not really been used for a few years but I am sure that it has had an influence on the program that is being used primarily for search result rankings now. 
link: https://www.semrush.com/blog/pagerank/#header2]]> 2021-02-25 23:16:30 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243610644 Recommendation Engine mvlroche https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243670394 Recommendation Engines are used in many places around the internet and other places that involve computers. Recommendation engines play a big role in how in internet is personalized for everyone. The engine is used to determine what a person may or may not like about a given topic. The math behind this engine is very simple because it uses the list of items and depending on what is selected makes predictions on what is liked and not, from this it can make more assumptions along the way as more items are selected. This can has a good impact because it allows for the scale of the internet narrowed down the user. ]]> 2021-02-25 23:51:18 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243670394 High Frequency Stock Trading mvcpoulleau https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243698889 To forecast price fluctuations, the financial industry has long used algorithms, but they are now being used in the burgeoning practice of high-frequency stock trading. This form of rapid-fire trading requires algorithms that can make decisions in the order of milliseconds, also called bots. By comparison, to both identify and react to possible dangers, it takes a person at least a full second. As a result, human beings are increasingly being cut out of the trade loop, and an entirely new digital world is emerging.

Link: https://io9.gizmodo.com/the-10-algorithms-that-dominate-our-world-1580110464]]> 2021-02-26 00:08:39 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243698889 sensor and navigation algorithm mvaspyrka https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243726661 This algorithm was made to control sensor and navigation systems on aircraft. It is fairly new, and tested first in Dec. 2020 on a U-2 Dragon Lady spy plane (the extremely high altitude spy plane).  Not much is known about this algorithm at the moment as the U-2 pilot and air force declined to say what the specific tasks are. To put it broadly, the algorithm was put in charge of the plane's radar senses and tactical navigation with very narrow duties. In fact,  it was said that this was the first time that a military system integrated AI, so obviously the pilot still had majority of the control as this is very fresh. While being trained against an opposing computer, the AI looked for oncoming missiles and launchers. For the in flight test, it got to decide where to direct the planes sensors. Of course, it will be a while before humans and AI team up on the battle field, but this allowed all of the branches to see the potential. I feel like the impact of this could be positive or negative, depending on your stance regarding war. Personally, I believe this will positively impact the air force pilots soon by adding an extra layer of security.

Link: https://www.washingtonpost.com/business/2020/12/16/air-force-artificial-intelligence/]]> 2021-02-26 00:24:43 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243726661 Middle Out Compression Algorithm mvsdhaliwal https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243735012 Middle out compression is something I first came across when watching HBO's Silicon Valley, but now Dropbox's Lepton is using the algorithm for real lossless image compression. Dropbox's Lepton uses a middle-out compression algorithm to reduce the size of JPEG-encoded images while being completely lossless. From the dropbox website "The JPEG format encodes an image by dividing it into a series of 8×8 pixel blocks, represented as 64 signed 10-bit coefficients. Thus the following 16×16 image would be encoded as 4 JPEG blocks.", this explains how the algorithm works to get this lossless compression. The impact of this code is that now images can be reduced without losing any of the original bits.
 Implementing Middle-Out compression here is the start of a wave of middle-out compression being used for all kinds of things from video chat to live streaming.  The code for this compression is open source and can be found on GitHub for anyone to play around with. 

Link to the article : https://techcrunch.com/2016/07/14/dropboxs-lepton-lossless-image-compression-really-uses-a-middle-out-algorithm/]]> 2021-02-26 00:29:47 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243735012 Algorithm in Video Games mvidacosta https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243738937 Algorithms in video games are used to control movement patterns and behaviors in a complex environment. Every time  the instructions of the steering algorithm are put inn place, a controlled object in the game moves. These video games use mathematical processes to mimic behavior that us humans have. For example, these algorithms use mathematical justification to figure out the exact amount of space or time that the character needs to make s certain move in a fighting game. I think that this will have a positive impact especially since the gaming industry is one that is thriving right now. Many people are interested in video games and the thought of another "world". By this I mean the complexity of another reality built because of these algorithms.  
 
Link to the article: https://towardsdatascience.com/how-a-chess-playing-computer-thinks-about-its-next-move-8f028bd0e7b1 

]]> 2021-02-26 00:32:12 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243738937 Algorithms in randomness mvrwilliams https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243829198 Randomness is apart of our daily lives, so it’s no surprise that it exists within the world of computer science too. Randomness can be in any way, shape or form. It can be numbers, letters, words, you name it. However, not all random number generators are truly random. Most random number generators use an algorithm to create a random number, and while this may be random enough, it isn’t truly random. People have found out how to create true randomness within the world of computers, using the most random thing of all, real life.

Link: https://www.howtogeek.com/183051/htg-explains-how-computers-generate-random-numbers/

]]> 2021-02-26 01:26:34 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243829198 Instagram Algorithms mvrzhai https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243863882 Instagram, the social media app used by millions around the world, uses an algorithm for their personal "Discover" page.  The posts or reels that are displayed on the discover page are geared towards what your activity on the app was or what you find the most interesting.  The main points of what Instagram looks at is:
1. Interest
2. Relationship
3. Timeliness
4. Frequency
5. Following
6. Usage
Most of the time, the more posts you interact with such as liking it or commenting on it. This leads to the relationship aspect of it. Instagram wants to prioritize posts from your friends and family. Another part is the timelines and frequency which calculates the number of times you open the app or even look at a post. In Instagram, there's another page called the discover page which delivers content that Instagram thinks you’ll be most interested in. Sometimes it's made up of content from accounts you already follow or to things you like.
Link: https://later.com/blog/how-instagram-algorithm-works/



]]> 2021-02-26 01:47:37 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243863882 TikTok FYP Algorithm mvimathers https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243917840 TikTok's FYP or "ForYou Page" is seen by many users and is the first page you see when you get onto the app. This page is generated to show videos similar to recent videos you've watched or to show you a certain genre of videos you regularly watch. The algorithm of this page is to real you in to stay on the app, if you keep seeing videos you're interested in as you keep scrolling down the ForYou page, the more you want to watch similar videos to the one previously shown. This algorithm keeps its users on the app and allows for more usage of the app from its consumers worldwide. This ForYou page also shows videos you've recently interacted with (liking, sharing, commenting), so if you've recently interacted with one of Charli D'amelio's dance TikTok's you're most likely to see her later again on your ForYou page.

Link: https://www.yrcharisma.com/2020/01/06/tiktok-fyp-algorithm/

]]> 2021-02-26 02:21:11 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243917840 Disney+ Algorithm mvmchow https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243998059 As one of the current largest and most popular streaming services, Disney+ has thousands of shows and hundreds of films, making it tough to decide and find what to watch next. To resolve this, Disney+ uses multiple algorithms that feed off behavioral data, making it smarter as time goes on. By studying people's program preferences, what they watch, don't watch, likes, and dislikes, the data can help decide which is the best algorithm that should be used for that user at the moment. The algorithms predict what a person might want to watch next and the relationship between various content to improve and personalize the user experience. The algorithms' positive impact is that the data allows Disney+ to understand their viewers better, enabling the service to make things more engaging for users. With the algorithms, the users can find what they desire to watch while also seeing and understanding what's available on the service and what it provides.

Link: https://www.forbes.com/sites/insights-teradata/2020/04/21/how-disney-plus-personalizes-your-viewing-experience/?sh=2bd76e4e3b6e ]]> 2021-02-26 03:16:28 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1243998059 Medicare Risk Algorithm mvjmates https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244018858 The Centers for Medicare and Medicaid Service (CMS) risk adjustment model uses the Hierarchal Condition Category (HCC) method to assess the different Medicare patients and calculate their risk scores. Certain medical diagnoses are put into groupings based on the use of resources that is required. The higher the anticipated healthcare costs, the higher the risk score. This system is used to adjust plan payments to ensure fair payment for providing healthcare and benefits for patients. These algorithms help to reduce the workload on office staff and to calculate an accurate cost for health plans. 
Link: https://www.foreseemed.com/medicare-risk-adjustment]]> 2021-02-26 03:31:33 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244018858 Tesla Autopilot Algorithm mvzmehmood https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244034822 The Tesla Autopilot system is one of the most advanced forms of self-driving artificial intelligence. The autopilot system uses algorithms to use the data that has been taken over hundreds of thousands of miles of using vehicles on the road and is able to process the data and calculate the path that a car should take. The data is compiled through Tesla's cars being driven and sending information through the network in addition to other sources of information that gathers data on different roads and highways. Using this data, the car will calculate which paths will have a risk of putting the car and passengers and danger and which is the safe path in addition to calculating what objects are cars and walls. Based on these calculations, the algorithm will plan out the best driving path and route for the car. This can have both a positive and negative impact, as some times the algorithm will work perfectly and the car won't have any problems, however other times the algorithm can cause problems and the car may need to be taken over by a human driver which can cause danger if the driver is not looking.

Link: https://digital.hbs.edu/platform-rctom/submission/machine-learning-the-engine-inside-teslas-automated-driving-technology/ ]]> 2021-02-26 03:42:47 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244034822 Google Search Algorithm mvbkim https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244049511 Google's search engine uses an algorithm to return relevant results to a user search. It takes information from search such as language inputed, specific information needed, and the meaning of the query to search for matching webpages. The webpages are analyzed for things like keywords on the page. Machine-learned algorithms also change over time to better match searches to results. The algorithm also attempts to prioritize reliable results with using a ranking system called PageRank and choosing sites that other users prefer for similar searches.

A possible negative impact may be biases in search results]]> 2021-02-26 03:53:30 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244049511 Job application algorithms mvalangbein https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244063832 Nowadays finding a residency or fellowship position/ program can be very difficult. So , in order to address this issue, the NRMP uses a mathematical algorithm to help match applicants to programs that would suit them best. It takes into account your location, qualifications, and your general needs. I have attached a link to a video better explaining the algorithm below:
https://www.nrmp.org/matching-algorithm/

]]> 2021-02-26 04:04:01 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244063832 BOA conversation algorithm https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244083748 The BOA app used an algorithm engine that can make an conversation with the user to help them search for what they want or help then to do things. The engine analyze the key word entered by the user and return relative result that can help you to narrow down you problem. I can compare the characters you entered in the search box with the services in the app to determine what the of question are you asking to give you a few choice that can help you to move on with your question. The good thing about it is that you can find answers for questions that you might not even sure about what you are looking for.

Link: https://promo.bankofamerica.com/erica/ ]]> 2021-02-26 04:19:22 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244083748 mvlroche https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244086240 2021-02-26 04:21:11 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244086240 The Euclidean Algorithm mvjyama https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244095508 The Euclidean Algorithm is an algorithm first described by the Greek mathematician, Euclid. It's purpose is to find the GCD (Greatest Common Divisor) between two whole integers. The importance of this algorithm stems from the fact that it is among the oldest algorithm devised that is still somewhat in use today. The algorithm was written by Euclid in the year 300 B. C. 

Source: https://www.khanacademy.org/computing/computer-science/cryptography/modarithmetic/a/the-euclidean-algorithm]]> 2021-02-26 04:28:44 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244095508 The google search algorithm mvabiswas2 https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244096470 The google search algorithm is an algorithm used to determine which websites show up in response to your search -- and in what order they show up. As an algorithm that affects Google, a search engine used by tons of people around the globe, it is very important. This algorithm is actually a whole group of algorithms to provide an article that has the answers you are looking for and is also high quality (has lots of information, is reliable, etc.) The specific words you use, the language you write the question in, your settings, your location, and even how easy it is for users with different devices or low internet speed to use this website all play a part, and more! 

This is an algorithm with an extremely good impact. In ranking the websites so that users can easily find what they are looking for within the first few links they click on, Google has it made it much easier for all it's users to find information when they are looking for it.

Of course, this all assumes that the algorithm is working properly. If there was in fact something wrong with the code that led it to not effectively find the best sites or be more biased towards/against some type of site, that would certainly be a bad impact - and it would affect a lot of people, too!

link: https://www.google.com/search/howsearchworks/algorithms/ ]]> 2021-02-26 04:29:34 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244096470 BOA conversation algorithm mvmsun https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244100975 The BOA app used an algorithm engine that can make an conversation with the user to help them search for what they want or help then to do things. The engine analyze the key word entered by the user and return relative result that can help you to narrow down you problem. I can compare the characters you entered in the search box with the services in the app to determine what the of question are you asking to give you a few choice that can help you to move on with your question. The good thing about it is that you can find answers for questions that you might not even sure about what you are looking for.
Link: https://promo.bankofamerica.com/erica/ ]]> 2021-02-26 04:33:29 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244100975 Akinator Algorithm mvamakram https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244106957 The Akinator website is an algorithm that asks general questions to try and find out what famous celebrity or character the user is thinking about. The engine analyzes your answers is slowly narrows down its options and questions that it asks the user. Finally, after asking around 10-20 questions, the engine reveals the character or person the user is thinking of. The impact is positive because the engine makes the game fun and cool as it is able to slowly find out the celebrity you are thinking about.
Link: https://en.akinator.com/game]]> 2021-02-26 04:38:06 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244106957 Instagram Algorithm mveeichten https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244113141 The Instagram algorithm decides which posts people see when they open their app. The Instagram algorithm has three primary ranking signals: Relationship, interest, and timeliness. Relationship means between the two accounts. Do you like their posts? Do you DM often? etc. Interest is if you have interreacted with posts like it in the past, you will be more likely to see this post. The more recent the post was, the more likely a person is to see it at the top of their feed. Obviously there are more important algorithm then if people see your instagrams posts or not. However, if you use Instagram to make money or promote your business and you don't know how the algorithm works, you are probably not reaching your full potential.
Website: https://blog.hootsuite.com/instagram-algorithm/#how]]> 2021-02-26 04:44:16 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244113141 Netflix Algorithm mvkraj https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244113361 The Netflix algorithm provides users of the streaming service a more personalized and entertaining viewing experience. The algorithm utilizes a variety of data such as viewer ratings/history, the time duration of a viewer watching a certain genre/show, the device the viewer is using, and the time of day the viewer is active. When a new user joins the website, it asks them what shows or movies they would like to watch, and the site then recommends shows and movies of a similar genre to those chosen by the user. Another interesting application the algorithm has on the site is the selection of thumbnails/artwork of suggested titles. For example, if the user watches a lot of action movies, an action scene from the movie will play if you hover over it, catching the user's attention. The algorithm has a positive impact because it provides a fresh and entertaining experience for the users.
source: https://medium.com/@springboard_ind/how-netflixs-recommendation-engine-works-bd1ee381bf81#:~:text=Netflix's%20machine%20learning%20based%20recommendations,and%20accurate%20the%20algorithm%20is.]]> 2021-02-26 04:44:30 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244113361 Netflix Algorithm mvjmoran https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244116007 Netflix uses machine based learning to curate a user experience based on the users interests. It does this by looking at nuanced plot threads and connecting them to similar content that the user likely wouldn't have otherwise chosen. This algorithm is thought of by many as the reason why Netflix is successful so it is highly important. ]]> 2021-02-26 04:47:16 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244116007 Twitter Algorithm mvjmakram https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244121695 Twitter, like other social media platforms, uses machine learning to sort out content based on ranking signals. The ranking signals include relevance, engagement, rich media, and recency. The algorithm calculates what you would like to see based on what you have liked in the past and what people like you have liked in the past. This leads to you mainly getting stuff you want to see. This can be both positive and negative. It can be positive by entertaining people but it can lead to people not making correct assumptions about certain topics. 


https://blog.hootsuite.com/twitter-algorithm/]]> 2021-02-26 04:53:04 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244121695 Among Us Algorithm mvdzilper https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244127586 In Among Us, the game determines who in the lobby will be the imposter, and which players will be the crewmates. For example, if the lobby settings are set to 1 imposter and 10 players, each player has a 10% chance of being the imposter in the game because the algorithm multiplies the number of imposters set for a lobby by 100, and then divides that by the number of players joining. This algorithm makes Among Us exciting and fun because no one knows who the imposter will be, but everyone has a fair shot at it. This makes everyone sus during the game because anyone can be the imposter.]]> 2021-02-26 04:58:36 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244127586 YouTube Algorithm mveakbarpour https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244133070 The YouTube algorithm gives the user videos to watch 70% of the time. And proven by Pew Research Center, 81% of YouTube users watch what is recommended to them by the YouTube algorithm.

At the start of YouTube's existence, for seven years, YouTube had a system where they rewarded video creators that got the most views instead of the ones that kept their watchers engaged. This led the video creators to use clickbait, which could mean misleading titles and thumbnails proliferated. The watchers would often get mad and feel tricked after clicking a video and watching something that wasn't what they expected, leading them to click off the video before finishing. Youtube realized this issue and changed its system many times after this. But still today, there are still video creators using clickbait.

Today's algorithm that YouTube uses is what they call a “real-time feedback loop that tailors videos to each viewer’s different interests.” It chooses the videos for each individual user. The algorithm tries to accomplish two things at the same time: find the right video for each viewer, and get viewers to keep watching. This means the algorithm is watching user behavior as well as video performance. The two main places that the algorithm affects are the search results and recommendation streams.

How the YouTube algorithm influences search results
Two different people could search for the same thing but get different results because of the algorithm. This is because of 2 reasons: 
  • Your video’s metadata (title, description, keywords) and how well those match the user’s query
  • Your video’s metadata (title, description, keywords) and how well those match the user’s query

How the YouTube algorithm influences recommended videos
First, the algorithm gives each video s score based on performance analytics data.

Second, it shows videos to people depending on what they have watched in the past and what people similar to them have watched.

The algorithm's goal is not to show the users "good" videos but rather videos they want to watch.
]]> 2021-02-26 05:04:10 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244133070 TikTok "FYP" algorithm mvjmathew https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244142615 The TikTok "FYP" or the For you page is a set of videos uniquely tailored to the user's interests. The system recommends content by ranking videos based on a combination of factors. The algorithm is based on the video you like, user interactions such as commenting and following, video information, and the user's account settings. Each of these factors are carefully weighted to create your personalized "FYP" This is part of the reason why TikTok is so popular among the newer generations because they find videos tailored to their interests and find people who are similar 

]]> 2021-02-26 05:13:59 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244142615 Strukturtensor Algorithm mvkrupp https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244156380 The strukturtensor algorithm is one used to process images, and divide pixels into different categories, such as pixels in regions with similar hues, pixels on the edge of a block of color, and those on a vertex. By splitting pixels into these groups, the algorithm is able to create a black and white image which shows the outlines of objects. In order to do this, it finds the differences between color values over a certain area. Different color differences correspond to different values for the new drawing. 
https://www.cs.cmu.edu/~sarsen/structureTensorTutorial/

]]> 2021-02-26 05:28:50 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244156380 Job Recruiting Algorithm mvazheng https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244164973 Job recruiting algorithms collect, scan, and rank the applicant's resumes for a job opportunity. Many large corporations use these algorithms to efficiently deal with thousands of resumes. Unqualified candidates can consistently and effectively be eliminated using job recruiting algorithms. The math behind the algorithm is that it looks for keywords and qualifications that can translate to a good candidate. In other words, the job recruiting algorithms will determine the strength of a candidate by their skills, which is based on the keywords featured in their resume. I think job recruiting algorithms are positive as long as it is balanced with human recruiting.
https://www.topresume.com/career-advice/what-is-an-ats-resume]]> 2021-02-26 05:39:31 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244164973 Facial Recognition Algorithm mvkouyang https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244166747 China is a world leader in facial recognition technology. The Chinese government has nearly all of it's 1.4 billion people in a facial recognition database. China also has roughly 200 million surveillance cameras. According to CNBC, Shanghai's YITU Technology can identify a face out of nearly 2 billion its database in just seconds (CNBC).

The algorithm detects faces using key facial features. It then uses statistics and other information such as location to shrink down the possibilities and select a final match. The impacts are both positive and negative. Big cities in China, such as Shanghai, are typically very safe due to the prevalence of facial recognition technology and police. It's hard to commit violent crimes if the authorities see almost your every move. On the other hand, these systems may invade the privacy of innocents who haven't done anything wrong. I personally think I would be okay with facial recognition technology, as long as the data wasn't be used for something else, because it would be what keeps me safe. In the future, this technology is likely to spread across the world and will hopefully help discourage crime in areas all over the planet.

https://www.cnbc.com/2019/05/16/this-chinese-facial-recognition-start-up-can-id-a-person-in-seconds.html ]]> 2021-02-26 05:41:50 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244166747 Algorithm mvashen https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244176899 One algorithm I'm interested in is the algorithm used to generate Tetris pieces in guideline Tetris games. The algorithm used is a certain kind of randomizer used to give the player consistent yet random pieces. The way the algorithm works is by grouping up the 7 Tetris pieces into groups of 7. Each group is called a bag, and the player is given consecutive bags of pieces. Each bag consists of one of every 7 Tetris pieces albeit in random order. This algorithm ensures a certain degree of randomness but also prevents the player from getting too many of the same piece consecutively or not getting a certain piece for too long. 

One interesting consequence of this algorithm is that players can use the properties of the randomizer to create patterns that can be built semi-consistently. This can also be used to create a pattern that guarantees that the player never dies - this pattern results in a clear board at the end and the pattern can be repeated indefinitely:]]> 2021-02-26 05:53:36 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244176899 Algorithmic Trading mvslim2 https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244179974 In stocks the market is very volatile, so it's quite hard to tell when and how the stock will either increase or decrease. By the use of algorithmic trading, the computer recognizes certain patterns of stocks that may occur when it's about to go up/down so they either buy or sell the said stock. This helps stock brokers make decisions faster without their judgement being clouded by emotions. The way this algorithm works is by identifying certain trends of the graph and then buying before the trend usually goes up and selling before the stock goes down. It uses statistics and the highest likely chance of trends succeeding and testing numerous different trends to further better itself. This not only benefits people who invest into their own stocks they prefer but also stock brokers who are helping other people make money as well. This is a positive impact as it helps people generate more wealth
https://www.quantstart.com/articles/Best-Programming-Language-for-Algorithmic-Trading-Systems/]]> 2021-02-26 05:56:50 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244179974 Cogniac Algorithm mvdmeyer https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244184692 The company Cogniac uses algorithms to recognize flaws in manufactured parts in order to identify weak points or cracks. It uses machine learning by being told what the thing it's looking for is so it can find it again even if it looks different. This can be used to find cracks by identifying what a proper part looks like and then comparing it to the part that's damaged.

https://cogniac.co/use-cases/rail]]> 2021-02-26 06:00:46 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244184692 Instagram Algorithm mvafotedar https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244197912 The Instagram algorithm usually goes through a cycle of things that are interest, timeliness, frequency, following, and usage. These are all things that add up to what will pop up on your Instagram feed. It is also based on old things you may have liked a long time ago. The more similar type of posts you like the more those types of posts will continue to pop there. In conclusion, the algorithm for your Instagram feed is basically a combination of all your Instagram behaviors and searches.
source:https://later.com/blog/how-instagram-algorithm-works/

]]> 2021-02-26 06:14:33 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244197912 Tesla's Autopilot Algorithm mvmturqueza https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244218963 The algorithm takes all of the data from the cameras and sensors around the car and combines all of the data to make an in-depth view of the world around the car. In order to train the neural networks to predict such representations, algorithmically create accurate and large-scale ground truth data by combining information from the car's sensors across space and time. This version of data collecting is not as in-depth of a view as LIDAR witch most if not all other self-driving car companies use Tesla’s way of detecting and putting the world around the car together allows it to operate in poor weather conditions such as rain. The LIDAR has a hard time seeing the world around it in poor weather conditions.

]]> 2021-02-26 06:35:01 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244218963 Google's Search Algorithm mvaverma https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244220614
This fascinating algorithm sorts through countless sites and directs users to the sites they are most likely to visit. While Google does not provide the math behind the algorithm, we are able to know some of the factors that determine the search results. First, the series of algorithms look at the input query from the user. It then searches for results using keywords and synonyms of the keywords in the query. It takes into account timeliness as well. For example, if you were to search for a trending topic or sports scores, the search results would be very recent because the user is likely searching for recent information. These complex algorithms are even capable of assessing what the user is searching for with only a few keywords. Instead of assessing relevance by counting how many times the article mentions keywords, the algorithm returns articles that specifically answer a query. The algorithm also takes into account the reliability and usability of relevant sites before the results are returned. In order to increase user-friendliness, the algorithm promotes websites that run quickly. Finally, the algorithm uses past user data in order to determine what site the user is most likely going to click on. The algorithms have been incredibly beneficial, giving access to knowledge to millions and allowing millions more to spread knowledge. The algorithms can potentially be harmful with the existence of intentional or unintentional biases. Because the algorithm promotes sites that are "trustworthy", it makes it easy for Google to potentially censor information. That being said,  so far Google has maintained incredible integrity with its practices.

Information from: https://www.google.com/search/howsearchworks/algorithms/]]> 2021-02-26 06:36:35 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244220614 Social media's Advertising algorithm's mvskan https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244243108 With the rapid expansion of social media's influence, a precise advertising algorithm is necessary to maximize profit. Social media websites collect user's data including videos you watch and posts you like. They then use this data to estimate what type of person you are. This estimate is auctioned off to different companies that appeal to your estimated type. Multiple factors contribute to a company's "value", these include how much text they have, or how much clickbait it has. Unfortunately, there isn't much information on the code of these advertising algorithms]]> 2021-02-26 06:55:53 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244243108 Tesla Autopilot/Autonomous Vehicles Algorithms mvhchaudhary https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244267993 Autopilot in vehicles is something that goes quite unappreciated. The complexity of Tesla's algorithms within autonomous cars is exceptional. The core algorithms used operate on the data coming from the many sensors and cameras placed around the vehicle. The data is sent into the CPU where the algorithms represent the surroundings in a true-to-size virtual space. This virtual space is subject to even more rigorous algorithms which break down the data and analyze the choices and best possible decision to fulfill while under autopilot. The algorithms act on real-world situations to determine an outcome. This is where the grey area of the algorithms is found. The real world is a place of uncertainty. Some of the decisions that the car is forced to make can be questionable while looking at it from a moral standpoint. If the car is moving on the highway and is forced to make an abrupt turn to avoid a truck near the front of the car. Does it swerve left where a van full of adults are driving, or into the helmet-less motorcyclist on the right? What if the motorcyclist was wearing a helmet, would that affect the decision? Overall, autonomous vehicles and the Tesla algorithms are extremely beneficial and have a huge positive impact. There will always be difficult situations that pose a threat to others or even yourself, but such a complex algorithm should not be invalidated due to a moral conflict.

Source: https://www.tesla.com/autopilotAI

]]> 2021-02-26 07:14:17 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244267993 YouTube's Recommendation Algorithm mvgyamaji https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244269260 The YouTube recommendation algorithm is one that has changed over time ever since its foundation in 2005. During the years of 2005-2012, YouTube used a system where they would push videos with the most views onto users' recommendation pages. However, this led to "clickbait" (misleading title and thumbnail) and user engagement was poor. YouTube then changed the system to user engagement in 2012. This new system promoted the message of "make videos that people want to watch". This allowed people's recommendation pages to be far more interesting and allowed for higher user experience quality. Then, from 2016-2020, YouTube started to demonetize videos that violated community guidelines as well as stop recommending videos that were "borderline" content (content that did not violate guidelines but was harmful and misleading). In the end, YouTube's recommendation algorithm has gone through some change and has allowed users to be able to watch the videos that they enjoy.

site: https://blog.hootsuite.com/how-the-youtube-algorithm-works/

]]> 2021-02-26 07:15:05 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244269260 Netflix Algorithm mvajacobsen https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244305084 Netflix uses machine learning, algorithms and creativity to find recommendations for you in order to accommodate the best they can to your interests. Netflix collects data from you such as the duration you spend watching a certain genre, what you watch after, what was watched before, what is watched right now. They also take time of day into consideration when giving recommendations. They categorize every movie, every show and choose the ones that best fit you. They then display it on your screen with the ones in the top left being most recommended because they are most likely to be seen while the ones in the bottom right are less likely to be seen. This algorithm is positively impacting the viewing audience because it always is updating and bringing new and interesting content to the viewer. 
Source: https://uxplanet.org/netflix-binging-on-the-algorithm-a3a74a6c1f59

]]> 2021-02-26 07:37:01 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244305084 Google Map Algorithm mvaroberts https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244344408 Google Maps uses an algorithm to find the shortest path to the users end destination. For example, when you type in the location of somewhere you would like to go, the algorithm searches for the most optimal, shortest and fastest distance in which it would take the least amount of time to get to the destination given. This is done by using Dijkstra's Algorithm. The algorithm is a greedy algorithm which attempts to find the shortest path between nodes on a graph. Google maps uses this algorithm to find the shortest path between destinations.

Source: https://medium.com/@yk392/dijkstra-algorithm-key-to-finding-the-shortest-path-google-map-to-waze-56ff3d9f92f0

]]> 2021-02-26 07:59:05 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244344408 Youtube's Reccomendation Algorithm mvekang2 https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244358071 Youtube's recommendation algorithm is extremely complicated, involving multiple equations and functions that the user's data is put through in order to create a set of recommended videos. Generally, there are two different algorithms that determine recommended videos. One algorithm compiles a large list of videos, while the other algorithm ranks the video, placing the most important data at the top of the feed. There is a lot that goes into this decision making, including harsh equations, rigorous preference monitoring, and popularity counters. An example of an equation is listed down below, where "i" is the item, "c" is the context, and where "u" is the user. The algorithm's complexity is what makes it so effective; however, it can still make non-perfect decisions from time to time.]]> 2021-02-26 08:06:42 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244358071 Youtube's Algorithm to keep users engaged mvcguo https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244375771 Youtube's Algorithm has changed a lot since the creation of the platform. Up until recently, (2016 - 2020) Youtube used a machine-learning algorithm to recommend and push videos to the users. According to youtube,  their current algorithm is a "real-time feedback loop that tailors videos to each viewer’s different interests". Youtube has not revealed the inner workings and calculations of its algorithm to prevent people from cheesing the system, but according to https://blog.hootsuite.com/how-the-youtube-algorithm-works/ (a very "legit" source that I found), Youtube's video algorithm takes in a variety of variables when recommending videos to its users.  Some of the variables that youtube takes into consideration include:

Amount of users clicking on the video
Watch time of each individual
Likes and dislikes
The age of the video

More user-based information like subscribed channels liked and disliked videos, what videos do the users ignore, and what channels do they engage most with are also included when calculating what videos they recommend to the user.]]> 2021-02-26 08:15:54 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244375771 Netflix Algorithm mvmyates https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244436193 The recommendation system for Netflix uses data collected from the person watching. Recommended shows are the result of viewing habits. Every time someone watches a movie or TV show, Netflix is collecting data for the algorithm. The algorithm keeps refreshing every time a person is active on Netflix and is constantly up to date. The algorithm takes into account the time of day you watch, what genres are being watched and for how long. Additionally, it sees if your viewing habits are similar to another Netflix watchers habits. In fact if you tend to watch the same things as another Netflix user, the algorithm will take note of this and recommend shows based on the other person’s viewing habits as well. Netflix wants to keep you engaged and interested to keep watching. They use the data collected and run it through the algorithm to show us shows and movies they think will intrigue us. Every few days a new thing to watch pops up in the recommended section for people to watch. 
Source: https://uxplanet.org/netflix-binging-on-the-algorithm-a3a74a6c1f59 ]]> 2021-02-26 08:45:41 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244436193 Euclidean Algorithm mvlbai https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244517545 Euclidean Algorithm is an algorithm that use used to find the great common factor (GCF) between two numbers. It uses the pattern when two number divide and has a remainder, the  denominator can be used as the numerator, and the remainder can be used as the denominator to write a new division calculation. These steps repeat until the remainder is 0. The GCF will then be the denominator of the final division with 0 remainder. Discovered by Greek mathematician Euclid, the Euclidean Algorithm has an positive impact by providing an efficient method to find the GCF between two numbers.

source: https://www.inchcalculator.com/euclidean-algorithm-calculator/

]]> 2021-02-26 09:26:05 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244517545 YouTube Algorithm mvsiyer https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244520922 The YouTube algorithm is, essentially, a real-time feedback loop that suggests videos based on a viewer's interests. The loop suggests such videos by examining a user's searches, previous videos watched, and user patterns. Ultimately, the algorithm has a couple goals it wants to accomplish: making sure the user is continuing to use its platform, and find the right video for the user.
The impact of the YouTube algorithm is positive. This is because it enhances the user/consumer experience through a simple, yet efficient, algorithm.

Source: https://blog.hootsuite.com/how-the-youtube-algorithm-works/#:~:text=According%20to%20YouTube%2C%20the%20algorithm,get%20viewers%20to%20keep%20watching.


]]> 2021-02-26 09:27:33 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244520922 League of Legend's matchmaking Algorithm mvskim https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244705859 As silly as it sounds, I've been recently getting into matches on this game called "league of legends", where I'd be paired up with one or two players with a low win rate in match-making (there are 5 people per team). This was causing me to lose more games then win even if my skills are great. Out of curiousity, I was able to figure out that League of legend's match making algrotihtm does this on purpose to balance out players winrates. If I, the user, have a winrate higher than 50%, I will be placed into a team with 1-2 players with winrates lower than 50%. On the otherhand, if the user has a winrate less than 50%, they will most likely be placed with a player with a high winrate. I can see why league of Legend's had decided to implement this algorithm, however, it gets annoying quickly knowing that you are losing more games due to your teamates. ]]> 2021-02-26 10:50:57 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244705859 PageRank - Google's mvzshabbir https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244813644 PageRank is an algorithm devised by Larry Page and Sergey Brin (the founders of Google) in 1998 while they were working on creating a new search engine at Stanford. PageRank was later utilized by Google Search to rank pages within its search engine results. Essentially, PageRank is a relative measure of the importance of a specific page on a site by counting the relative number and quality of links connecting to a page to determine the relative importance/significance of said page. The more important the PageRank algorithm deems the webpage, the higher it appears in Google's search results. Although it is no longer the only algorithm used by Google to filter search results, it was the first algorithm that was used to do so and is the most widely known as a result of it being very unique for its time. This algorithm has had significant positive and negative effects on the world, as it ushered in a new era of search engines and technology, being responsible for Google's success, one of the largest companies in the world. Unfortunately, today there is much controversy over the power of Google in the lives of billions of people around the world, which is a negative effect of the PageRank algorithm. The math behind the PageRank algorithm is explained in the image below.

Source:
https://interestingengineering.com/15-of-the-most-important-algorithms-that-helped-define-mathematics-computing-and-physics]]> 2021-02-26 11:42:25 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1244813644 Netflix Recommendation Algorithm mvsbhasin3 https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1245554049 Netflix has been making sure that its algorithms highlights their newest shows, as well as some older shows and movies that they think a consumer will enjoy based off of the users past viewing history, along with their location. For example, say a user watched more romantic comedies, and lives in Japan. The Netflix algorithm will be sure to recommend more romantic comedies, like the ones they have liked in the past, that are in Japanese, to the user. This algorithm ensures that consumers will intake new shows and movies constantly, and this results in exposure to new content. Strong recommendation algorithms also boost viewership on their platform. By their titles that attract public attention, alongside their marketing being on numerous media platforms, they are guaranteed a viewer spike in the first week when a show comes out due to fueled interest that can start months before a show or movie is even out. Furthermore, they make sure to put shows and movies on your content library that they assume you will like due to your viewing history, which, in turn, has an outcome of more clicks, or really just spikes in views of their content. This corporations estimates that about 75% of their viewership is based off of recommendations too. Last December, Netflix decided to remove their region-based algorithm in their recommendation system because of their current efforts for global expansion. Because of this, they are making shows in multiple languages, which has currently broadened the choices the recommendation algorithm can recommend to users of all kind.
Source:
https://www.martechadvisor.com/articles/customer-experience-2/recommendation-engines-how-amazon-and-netflix-are-winning-the-personalization-battle/#

]]> 2021-02-26 15:13:28 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1245554049 Youtube Algorithms - Arjun Bhandari https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1245701983 Since there are over 2 billion users on Youtube, how does the algorithm show people what they want to see? When Youtube started in 2005, the algorithm favored clicks overwatch time. This means that if a person clicked on the video, the algorithm would pick up that the user was interested in the topic. But some problems with that were the click-baiting and putting false titles. Youtube fixed the algorithm in 2012, when they decided to use the watch time - how long the user spends on a video - to decide what videos to recommend. This was a lot better, and it helped deliver better recommendations to the users. Finally, in 2016, Youtube released the machine learning part of the algorithm, (shown in the picture below), demonstrating the funnel that the algorithm filters through to recommend the right videos to the right people.

Source: https://blog.hootsuite.com/how-the-youtube-algorithm-works/


]]> 2021-02-26 15:44:39 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1245701983 App Store Ranking Algorithm mvarock https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1245713998 The full details of how exactly the ranking of apps from a search is made is unknown but most of what occurs is either fairly obvious or can be inferred. For example when you enter a query in the App Store, it tries to pull together apps with the same or similar/relevant names. For example if you were to type "news", apples' news app might appear alongside other media sources. If you were to type "noticias", Spanish news media would appear. Apps with relevant or similar names are sorted on the basis of Downloads, Ratings, Keywords, and updates. The popularity off an app will determine where it is. Someone is much more likely to find and be interested in a 5-star app with 300,000 reviews over its 2-star, 2 review counterpart. The subtitle of an app might also lead your search for it. If the potential query "Made for awesomeness" is created, then in the list of apps that appear, some may have the subtitle "Made for awesomeness". Apps are also sorted based on how recently and frequently they update. Most people aren't looking for some outdated food delivery service no longer in business from 10 years ago, they only worked in certain areas and were pretty small compared to current business anyway. You either want door dash or some kind of equivalent because they are relatively modern apps that are actually still being used, and if you were to order something the app would work whereas one from 10 years ago no longer in business may not.



]]> 2021-02-26 15:47:18 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1245713998 Reddit Algorithm - Humza Mahmood mvhmahmood https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1245799852 Reddit is a website known as a message-board with a page for trending posts. Their algorithm was created to show only the best and most relevant posts to users. They do this by calculating the amount of upvotes, or likes, a post has relative to the amount of time it has been up. Furthermore, the Reddit algorithm also calculates a percentage of user engagement, which simply means that Reddit calculates what percent of users who see the post interact with it. This algorithm is designed to only show posts to users that are engaging and will keep the user on the platform for as long as possible.

Reddit also personalizes the posts that people see. Depending on what types of posts the user has engaged with in the past, Reddit will try to show posts which are similar in nature to keep the user engaged.

Image source: https://medium.com/hacking-and-gonzo/how-reddit-ranking-algorithms-work-ef111e33d0d9]]> 2021-02-26 16:05:21 UTC https://padlet.com/banwar1/cmz5k27huynx9kdf/wish/1245799852