OPTIMIZING STOCK TRADING STRATEGY WITH
K-MEANS CLUSTERING
Mini Project
Big Data Analytics[BDA]
Department of Computer Engineering
TERNA ENGINEERING COLLEGE
Nerul (W), Navi Mumbai 400706
FINAL PRESENTATION
Group Members:
Amey Thakur B-50
Under the Guidance of :
Hasan Rizvi B-51
Prof. D. M. Bavkar
Mega Satish B-58
INDEX
1. Introduction
2. Literature Survey
3. Problem Statement
4. Proposed Methodology
5. Model Deployment
6. Snapshots
7. Conclusion
References
ABSTRACT
➔ We propose to determine a single exchange-traded fund (SPY) investing strategy that
will maximise our total wealth.
➔ We intend to forecast future price fluctuations for a specific stock.
➔ We develop a predictor for multiple firms using these trained RL models that forecasts
the every day close stock prices.
➔ The goal of this project is to learn and get hands-on experience in Data Analytics and
Machine Learning.
PROBLEM STATEMENT
➔ The main emphasis and objective of our project is to analyse given raw data and do
exploratory data analysis in order to fully comprehend and identify patterns.
➔ Then, using a Neural Network approach, construct a model and train it to get the
desired outcomes.
➔ Finally, it will be deployed as a web application.
LITERATURE SURVEY
To get a better understanding of K-Means Clustering and its uses for the stock trading strategy, we
referred to the following papers:
➔ “Fast K-Means Clustering for Very Large Datasets Based on MapReduce Combined with a New
Cutting Method”, by Duong Van Hieu and Phayung Meesad. This paper gives an overview of the
importance of the K-Means Clustering Algorithm for Big Data. This paper presents a new
approach for reducing the number of iterations of the K-Means algorithm which can be applied
to very large dataset clustering.
➔ “Enhancing stock prediction clustering using K-means with genetic algorithm”, by E. N. Desokey,
A. Badr and A. F. Hegazy. The main objectives of this paper are to optimize the clustering of
stock market prediction and to examine the impact of applying genetic algorithm optimization
with the k-means clustering algorithm. The evaluation shows that using genetic algorithm and
k-means clustering algorithm with Chi-square similarity measure achieved the highest accuracy
with the least sum of square distances.
INTRODUCTION
➔ We've always been captivated by the stock market's seeming unpredictability. So, for a
day trader, deciding what to trade is the first and most important step.
➔ Day traders should look for equities with plenty of liquidity, moderate to high volatility,
and a large number of followers. Isolating the present market trend from any
surrounding noise and then capitalising on that trend is the key to finding the
appropriate stocks for intraday trading.
➔ It's critical to devise a well-balanced, low-risk plan that will benefit the majority of
individuals. One such technique that we propose uses K-means clustering to generate
automated trading strategies based on previous data.
METHODOLOGY
K-Means Clustering
➔ To process the learning data, the K-means algorithm in data mining starts with the
first group of randomly selected centroids, which are used as the beginning points for
every cluster, and then performs iterative (repetitive) calculations to optimize the
positions of the centroids.
➔ It halts creating and optimizing clusters when either:
◆ The centroids have stabilized — there is no change in their values because the
clustering has been successful.
◆ The defined number of iterations has been achieved.
MODEL DEPLOYMENT
The necessary files for model deployment are the following:
1. app.py
2. requirement.txt - Contains a list of all the dependencies that your code requires in
order to function properly.
3. Procfile - In an app, a Procfile is a list of process types.
4. setup.sh
SNAPSHOTS
SNAPSHOTS
REFERENCES
1. Segmentation of stock trading customers according to potential value (2004)
https://doi.org/10.1016/j.eswa.2003.12.002
2. Customer Profitability Analysis, Cost System Purposes and Decision Making Process: A Research Framework
3. https://www.researchgate.net/publication/325092899_Customer_Profitability_Analysis_Cost_System_Pur
poses_and_Decision_Making_Process_A_Research_Framework
4. Van Hieu D., Meesad P. (2015) Fast K-Means Clustering for Very Large Datasets Based on MapReduce
Combined with a New Cutting Method. In: Nguyen VH., Le AC., Huynh VN. (eds) Knowledge and Systems
Engineering. Advances in Intelligent Systems and Computing, vol 326. Springer, Cham.
https://doi.org/10.1007/978-3-319-11680-8_23
5. E. N. Desokey, A. Badr and A. F. Hegazy (2017) "Enhancing stock prediction clustering using K-means with
genetic algorithm," 2017 13th International Computer Engineering Conference (ICENCO), pp. 256-261,
doi: 10.1109/ICENCO.2017.8289797.
THANK YOU
