View Proposal

---

Proposer

Habte Lejebo Leka

Title

Stock Price Prediction and Analysis System

Goal

Description

The project aims to develop a web-based stock price prediction and analysis system using historical market data provided in the resources section. The system will help users to explore historical stock performance, visualize relevant market indicators. It also helps users generate short-term stock price predictions using suitable data-driven modelling techniques. The project will involve preprocessing historical financial data, performing exploratory and time-series analysis, identifying and engineering relevant features, developing predictive models, and evaluating their performance using appropriate prediction metrics. The developed system will provide an interactive web-based dashboard for visualizing historical stock prices, market indicators, predictions, and model performance results. The students under this project are expected to investigate and justify an appropriate predictive modelling methodology for the selected problem. The choice of methodology should be supported by relevant literature and the characteristics of the available data. Students should conduct appropriate experiments, compare the performance of their developed models where applicable, and critically discuss the results, limitations, and potential improvements.

Resources

"Dataset: Stock Market Dataset Kaggle Link: https://www.kaggle.com/datasets/camnugent/sandp500 Leka, Habte Lejebo, et al. ""A Hybrid CNN-LSTM Model for Virtual Machine Workload Forecasting in Cloud Data Center."" 2021 18th International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP). IEEE, 2021: 474-478."

Background

Url

Difficulty Level

Moderate

Ethical Approval

None

Number Of Students

2

Supervisor

Habte Lejebo Leka

Keywords

stock price prediction, stock market analysis

Degrees

Bachelor of Science in Computational Sciences and Software Engineering