Financial Market Analysis Platform for Turkish Stock Market Investors

An AI-powered financial market analysis platform for Turkish stock investors; offering real-time news, stock predictions, company insights, technical analysis tools and comparison between past and today, in terms of finance.


hero-image shape shape
Features

Main Features of InvestmentHelper-AI

Investment Helper is designed for easy and seamless use of AI in investing, offering features that simplify investment decisions and enhance the overall experience for investors.

AI Chatbot for Company Analysis

Ask any questions about finance that is related with some selected companies of BIST

Real Time News Analysis & Notifications

Receive instant notifications when something important to you happens about the stocks that you are interested in.

AI Based Stock Predictions

Check what machine learning algorithms tell for the future of selected stocks.

Technical Analysis Helper Toolkit

Do technical analysis on stocks with ease.

Comparison of Today and Past for Stocks

Compare past news and stock trends with today's conditions to gain insights into future trends.

About the Project

Bilkent University - CS491 Course Project

InvestmentHelper-AI is a project that is done for the CS 491 Course of Bilkent University. We plan to create a platform for the Turkish Stock Market investors that will help them while taking actions. Our supervisor for the project is Asst. Prof. Özgür Salih Öğüz and innovation expert is Murat Kalender.

The essence of the project is to provide users with a comprehensive finance companion. The companion will include a chatbot, that the users can interact with and ask comparative questions about the past and current state of the market. Also, through the chatbot, the users can enter the context of the news that they want to be notified about in real-time (e.g “Tell me when interest rate changes.”), then, financial news from the web will be analyzed and relevant notifications will be sent to users when detected. The live analysis of financial news, that will allow us to notify the users such that the users will be able to have an initiative regarding their financial portfolio, will be established through advanced Time-Dependent Knowledge Graph based RAG pipelines utilizing LLMs. The news that will be used will be scraped from the web, from reliable sources of finance. The companion will also provide stock predictions based on sentiment analysis of financial news and past numerical data, recognize current trends based on current and past context of the market, and allow technical financial analysis on graphical representation of various stocks. The tool will utilize artificial intelligence for the functionalities defined above.

about-image
Reports

Our Reports in DOCX and PDF Format

These reports are prepared for the CS491(Senior Design Project) course.

Project Information Form - Oct 18

  • Name of the project
  • Description of the project
  • Supervisor & Innovation Expert
  • URL of the project web page

Meeting 2 Presentation - Nov 12

  • Web scraping
  • Translation
  • Market analysis
  • Other technical details

Assessment of Innovation Form - Oct 25

  • Innovation analysis
  • Unadressed risks for implementation
  • Suggestions for the project
  • General review of the innovation expert

Project Specification Document - Nov 22

  • Brief project description and requirements
  • Constraints: economic, social, ethical
  • Engineering standards and ethical responsibilities
  • Feasability discussions

Analysis and Requirements Report - Dec 13

  • Current system
  • Proposed system
  • System models
  • Other analysis elements
shape
Further Information

Looking for details?

Some detailed explanations to critical parts of the project

We plan to compare the latest trends of a stock with it's past trends. Also, we will take into account the news and financial environment of the past(and compare it with today's) and try to find the closest time interval in the past, compared to today. By doing that, we plan our users to take a look at the past, to have a more clear idea about the future.
CS491 Senior Design Project I course is the first half of two-semester course that will allow all senior CS students to gain and practice the required abilities and skills in the engineering design process. All students enrolled in the Computer Engineering program at Bilkent University have the opportunity to learn and practice engineering design, through their coursework and, CS491 and CS492 capstone project courses will give a unique opportunity to demonstrate these skills within a team of 3-5 students. CS491-CS492 projects are directed toward an innovative solution to a real and substantial problem. Students need to show their engineering design skills to solve practical and theoretical engineering problems that they will choose and pre-approved by the faculty. Below you can see the key outcomes of this course: Students will gain ability to identify, formulate and find innovative solutions to the open-ended engineering problem that is chosen and will translate their computer science knowledge into new context to design and implement an innovative solution as their capstone project. Students will experience collaborative work environment and will work productively as an individual and in a team on a project that produces a innovative solution to selected problem. Students will improve their project and team management skills, improve the ability to follow directions, instructions and project plans, and improve the ability to organize and use time effectively. Students will show their ability and competency in oral and written communications skills through oral presentations during project presentation, course seminars and technical reports. Students will learn to incorporate appropriate engineering standards while culminating a major engineering design experience which is based on the knowledge and skills acquired in earlier course work. Students will demonstrate and improve their ability to recognize ethical and professional responsibilities, global, cultural, societal, economic and environmental constraints in engineering while implementing their capstone project. Students will learn to do basic budgeting & feasibility Analysis for a start-up idea Students will learn ethics and intellectual property considerations while building a new product or service and learn how to do market competition analysis. Progress and final output of teamwork needs to be presented in both written and oral form during the course. Teams are advised to select complex and challenging subjects so that, they can demonstrate the knowledge and skills that they got from the engineering coursework, to analyze, develop, test, iterate, improve and deliver their design concepts using modeled or simulated artifacts and working prototypes.
We plan to create a Chatbot by using advanced Retrieval Augmented Generation techniques. The chatbot will remember all the past news and other financial events about some companies(selected by us). When a user asks a question, the chatbot will return the up to date information unless the user asks about something in the past. Also, we plan to create a chatbot which is capable of answering multi-hop questions that needs different informations from different news-events.
We are going to analyze finanical news and events in real time and notify the users if something important happens. Also, users will be able to enter what kind of events that they are interested in by entering natural language prompts. So, they will be notified whenever something happens that is important for them.
We are going to make stock predictions using the past stock prices & financial news. We plan to train a machine learning model using the state of the art algorithms and provide predictions to our users.
We will include the Technical Analysis Helper Toolkit, which allows the users to visualize some stock data with different types of charts and perform different kinds of technical analysis in a simple and intuitive way. AI will be injected into that toolkit, which will automatically detect trading patterns, including-but not limited to-the head-and-shoulders pattern, double tops, and more, that will help investors make educated decisions.
Our Team

Meet The Team

We are a group of 3 senior year Computer Science students.

team
shape shape
Hakan Muluk
hakan.muluk@ug.bilkent.edu.tr
team
shape shape
Mehmet Onur Özdemir
mehmetonur@ug.bilkent.edu.tr
team
shape shape
Emir Şahin Dilli
sahin.dilli@ug.bilkent.edu.tr
CONTACT US

Thank you for visiting,
love to hear from you!

How Can We Help?

hakan.muluk@ug.bilkent.edu.tr

sahin.dilli@ug.bilkent.edu.tr

mehmetonur@ug.bilkent.edu.tr