Group project details:

1. Please identify the dataset and put the link here.

2. State the initial intention of your project (simple description of your group's plan)

3. List the work distribution among group members;

   e.g., Nur Intan (student ID) - smoothing method experiment and discussion part.

Please state any questions/confusion that need justification and clarification here (through you group's column)

1. Dataset link:-https://www.kaggle.com/datasets/ihelon/coffee-sales/data

2. Initial intention of the project:-

   To optimize inventory refill through customer behaviour analysis. Analyse customer preferences by time and purchasing behaviour to improve inventory refills automation.

3. List the work distribution among group members;

   Joyce Lim Xinjie (23101205) - Literature Review, Experiment and Analysis (Model A)

   Tee Li Hong (23201604) - Project Background part, Methodology part

   Teh Ming En (24102225) - Abstract, Experiment and Analysis (Model B), Conclusion, Poster

1. Dataset from kaggle:

https://www.kaggle.com/datasets/aslanahmedov/walmart-sales-forecast/data

2. Build a time series forecasting model to predict weekly sales at the store and department level. The goal is to support data-driven decisions in inventory, staffing, and promotion planning.

3. Task Distribution:

Ang Pei Ying (23202671)

-Related Work, EDA, Use Case

-Model 1

Goh Deng Vee (23101176)

-Problem Statement, Introduction, Objective

-Model 2

Low Jia Kai (23102165)

-Abstract, Future Work, Conclusion

-Model 3

1. We got the data from Kaggle: https://www.kaggle.com/datasets/aslanahmedov/walmart-sales-forecast/data. There are 4 CSV files total: features, stores, test, and training datasets.

2. The main goal of this project is to build a machine learning model to forecast weekly sales for Walmart stores across 45 locations. The objective is to predict future weekly sales using past data while understanding how holidays and seasons affect sales. Achieving this will help Walmart make better decisions and improve business strategies.

3. Members

   LEE CHONG KEAT - 23201641

   • Project Background

   • Methodology - Project Framework

   • Methodology - Feature Engineering

   • Experiment and Analysis - Modeling 1

   • Deployment the code to dashboard

   LOW JUN HOE - 23201640

   • Literature Review

   • Experiment and Analysis - Modeling 2

   • Evaluation and Result Analysis

   Liang GuangTong -

   • Methodology - Dataset Preparation

   • Methodology - Exploratory Data Analysis

   • Experiment and Analysis - Modeling 3

   • POSTER

1. Dataset Confirmation and Link:
The dataset used in this project contains official daily price information of major vegetables and fruits in Nepal, including the highest, lowest, and average selling prices.
Dataset link: Time Series Price Vegetables and Fruits

2. Project Motivation (Research Plan):
This project aims to analyze historical price data of vegetables and fruits in Nepal and forecast future price trends. We will apply time series analysis and machine learning methods to explore seasonal patterns and price fluctuations. The goal is to provide data-driven support for dynamic pricing strategies in agricultural markets and e-commerce platforms.

3. Group Members and Task Allocation:

• XIA ZHIHAN – Responsible for data cleaning, visualization, and implementing smoothing methods (e.g., moving average, exponential smoothing)

• GAO HUAN – Responsible for time series modeling and forecasting (e.g., ARIMA, Prophet), model evaluation, and writing the project report

Dataset: stores_sales_forecasting.csv (2121 records, 2014-2017 data)

Use Case: Predicting daily sales in a retail store to optimize inventory management and reduce over-stock or out-of-stock situations.

Group Members and Task Allocation:

WAN JUN -Prophet model, including model implementation and parameter adjustment, seasonal analysis, visualization of prediction results.

SU KUNYUAN-Description of the implementation background, and corresponding project goals, Exploratory Data Analysis (EDA) : mainly including, data preprocessing and cleaning, time series visualization, trend and seasonal analysis.

https://www.kaggle.com/datasets/michealknight/personal-ecommerce-website-ad-cost-and-viewer-count/data

1. Goal: To develop a time-series forecasting model to predict daily website page views. The model will incorporate advertising spend and calendar variables (example holidays, weekends) to support smarter marketing and budgeting decisions.

Dataset: Corporación Favorita Grocery Sales Forecasting
🔗 https://www.kaggle.com/competitions/store-sales-time-series-forecasting/data

2. Initial Intention of the Project:

This project aims to forecast daily grocery sales across multiple stores using time-series analysis techniques. By leveraging historical sales data along with contextual information such as promotions and holidays, we intend to analyze demand patterns and provide insights for better inventory and retail planning.

3. Work Distribution Among Group Members:

• WU XIAOFENG（23305440） – Responsible for data preparation, feature engineering, and contributing to the methodology and conclusion sections.

• HE YULIN （23202484）– In charge of exploratory data analysis, visualization, and literature review.

• WANG ZEYU（23202225） – Handles experiment setup, result evaluation, and project background writing.

1. Dataset Identification (with link):

Dataset: Retail Store Inventory Forecasting Dataset

https://www.kaggle.com/datasets/anirudhchauhan/retail-store-inventory-forecasting-dataset

2. Initial Intention of the Project:

The goal of this project is to use historical sales data to predict retail inventory needs, thereby achieving better inventory management and operational efficiency.

3. Work Distribution Among Group Members:

WU XIAOFENG(23305440)

- Abstract

- Project Background

- Literature Review

- Experiment and Analysis(Model A)

HE YULIN(23202484)

- Project Framework

- Dataset Preparation (Preprocessing)

- Experiment and Analysis(Model B)

- Poster

WANG ZEYU(23202225)

- Exploratory Data Analysis (EDA)

- Feature Engineering

- Experiment and Analysis(Model C)

- Conclusion

1. Dataset:

We will use the Store Sales – Time Series Forecasting dataset from Kaggle.

Link: https://www.kaggle.com/competitions/store-sales-time-series-forecasting/data?select=train.csv

2. Initial Intention of the Project:

Our project aims to predict the future daily sales of products across different store locations in Ecuador using time-series forecasting models. We will explore and compare three different forecasting models (Prophet, ARIMA, and LSTM/XGBoost) to evaluate their performance and applicability in real-world retail sales planning.

3. Work Distribution:

• Member A (Name + Student ID): Responsible for 1.0 Project Background, poster preparation, and Model 1 (Prophet).

• Member B (Name + Student ID): Responsible for 3.0 Methodology, data preprocessing/cleaning, report formatting, and Model 2 (ARIMA).

• Member C (Name + Student ID): Responsible for 2.0 Literature Review, 4.0 Experiment & Analysis, 5.0 Conclusion, and Model 3 (LSTM or XGBoost).

LU XIAOLIN 23202992

SUN TAO 23202077

1.Dataset:https://www.kaggle.com/datasets/huthayfahodeb/top-10-cryptocurrency-price-data

This Dataset contains historical price data for 10 cryptocurrencies spanning from 2021 to 2024, in three different time frames: 1 day, 4 hours, and 1 hour. The data is sourced from the Binance API and stored in CSV (Comma Separated Values) format for easy accessibility and analysis.

2. We choose Topic3 Financial Time-Series Forecasting

Initial intenion of our project:

Build a time-series forecasting framework to predict future cryptocurrency prices for the top 10 most traded cryptocurrencies.

The project aims to support data-driven decision making in portfolio optimization, risk management, and dynamic trading strategies.

By forecasting trends in highly volatile crypto markets, the project helps investors and financial analysts to better manage investment risks and optimize return strategies.

3. Work distribution

   Member 1:

• Literature Review, Exploratory Data Analysis (EDA), Use Case Design

• Model 1

Member 2:

• Problem Statement, Introduction, Project Objectives

• Model 2

Member 3:

• Abstract, Future Work Suggestions, Project Conclusion

• Model 3

1) https://www.kaggle.com/datasets/manjeetsingh/retaildataset/data

2) Build a time series forecasting model to predicts sales for retails using historical retails data which covers from 2010-02-05 to 2012-11-01.

3) team task distribution

1) Nursyaza Nisa Binti Arfarizal (23203142)

-Objective

-Literature Review on domain and model 1

-Experiment and Analysis Model 1

-EDA 1

2) Siti Hanis Syafiqah Binti Hamid (23203058)

-Background of the Problem Domain

-Issue and Problem Statement

-Literature Review on domain and model 2

-Experiment and Analysis Model 2

-EDA 2

3) Nur Maisarah Binti Zulkifli (23203088)

-Literature Review on domain and model 3

-Experiment and Analysis Model 3

-EDA 3

-Future Work & Conclusion