See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/381956007

Analysis of Stock Price Predictions and Forecasting using Machine Learning

Article · June 2024

CITATIONS READS
0 138

3 authors, including:

Ijeasm Journal
International Journal of Engineering Applied Science and Management
53 PUBLICATIONS 1 CITATION

SEE PROFILE

All content following this page was uploaded by Ijeasm Journal on 11 July 2024.

The user has requested enhancement of the downloaded file.

 International Journal of Engineering Applied Science and Management
ISSN (Online): 2582-6948
Vol. 5 Issue 6, June 2024

Analysis of Stock Price Predictions and Forecasting using Machine

Learning

Priyanka Prasad1, Megha Mishra2

M.Tech Scholar, SSTC, Bhilai, CG1
Associate Professor, SSTC, Bhilai CG2

Abstract: This empirical research explores the application of machine learning (ML) algorithms for
stock price prediction and forecasting in the context of the Indian share market. The primary objective is
to assess the effectiveness of ML models, including Long Short-Term Memory (LSTM), Support Vector
Machines (SVM), Random Forest, and XGBoost, in predicting future stock prices based on past trends.
Historical data is sourced from reputable financial data providers, ensuring comprehensive coverage of
market dynamics. The dataset comprises 12,600 daily data points (5 years * 252 trading days/year per
company), segmented into training (70%), validation (15%), and testing (15%) sets. Data preprocessing
involves cleaning, handling missing values, outlier detection, and scaling for improved model
performance. Feature engineering techniques, such as creating technical indicators (e.g., Moving
Average Convergence Divergence (MACD), Relative Strength Index (RSI)), enhance predictive
capabilities. Each ML algorithm is trained on the training set, with hyper parameter tuning optimized
for accuracy and robustness. Evaluation metrics like Mean Squared Error (MSE) and Root Mean
Squared Error (RMSE) are computed on the validation set to identify the most effective models. The
final models are validated on the testing set to assess generalizability and real-world applicability. The
study reveals that XGBoost outperforms other algorithms with an accuracy of 86.5%, demonstrating its
efficacy in predicting stock prices. LSTM follows closely with 82.3% accuracy, highlighting its
capability in capturing long-term dependencies in time series data. SVM and Random Forest achieve
accuracies of 79.6% and 85.1%, respectively, showcasing their competitive performance. Feature
importance analysis identifies open and close prices as critical predictors across all models,
underscoring their influence on future price movements. The findings suggest that ML algorithms,
particularly XGBoost and LSTM, offer promising avenues for enhancing stock price prediction accuracy
in the Indian share market. These insights can inform investment strategies and risk management
practices, contributing to more informed decision-making in financial markets. Future research could
explore advanced ML techniques and incorporate additional market indicators to further refine
predictive models.

Keywords: Stock price prediction, Machine learning, Indian share market, Long Short-Term Memory
(LSTM), XGBoost, Feature engineering, financial forecasting.

1. Introduction foundational but often struggle to capture the inherent

volatility and unpredictability of stock prices. This
Stock market forecasting is a critical endeavor for investors limitation stems from the dynamic and non-linear nature of
and traders seeking to navigate the complex dynamics of financial markets, where various external factors, investor
financial markets. Traditional methods of forecasting, such sentiments, and global events can swiftly influence stock
as technical analysis and fundamental analysis, have been prices.

Paper ID: 2024/IJEASM/5/2024/2089 1

 International Journal of Engineering Applied Science and Management
ISSN (Online): 2582-6948
Vol. 5 Issue 6, June 2024

Machine learning (ML) has emerged as a promising traded companies. Each dataset will encompass daily
alternative in stock market forecasting due to its ability to opening, closing, high, and low prices over the past five
analyze vast amounts of historical data and identify years, alongside pertinent time series data such as trading
complex patterns that may not be apparent through volume and market indices. This robust dataset will consist
traditional methods. Unlike rule-based systems, ML models of 1260 daily data points per company, resulting in a total
can adapt and learn from data, thereby potentially of 12,600 data points across all companies (Section 3.2).
improving prediction accuracy over time. This ample data volume ensures sufficient depth for
One of the popular ML techniques applied to stock market training and validating the machine learning (ML) models,
forecasting is Long Short-Term Memory (LSTM), a type of with the dataset partitioned into training (70%), validation
recurrent neural network (RNN) designed to process and (15%), and testing sets (15%) to enable thorough model
predict time series data with long-term dependencies. evaluation.
LSTM models excel in capturing patterns in sequential The study will employ a variety of established ML
data, making them suitable for analyzing stock price algorithms for stock price prediction, including Long Short-
movements that exhibit temporal dependencies and Term Memory (LSTM), Support Vector Machines (SVM),
irregular patterns. Another effective ML approach is Random Forest, and XGBoost. Each algorithm's
XGBoost, an ensemble learning method known for its effectiveness will be assessed through rigorous data
scalability and performance in regression tasks. XGBoost preprocessing techniques, encompassing cleaning, handling
combines the predictions of multiple weak learners missing values, outlier detection and removal, and data
(decision trees) to achieve higher accuracy, making it scaling to enhance model performance. Feature engineering
valuable for forecasting stock prices based on a wide range will also be conducted to create new features from existing
of input features. data, potentially incorporating technical indicators like
Feature engineering also plays a crucial role in enhancing Moving Average Convergence Divergence (MACD) and
the predictive power of ML models in stock market Relative Strength Index (RSI). Model training will involve
forecasting. Engineers can create additional features from optimizing hyperparameters to maximize predictive
raw data, such as technical indicators (e.g., Moving accuracy on the validation set, utilizing metrics such as
Average Convergence Divergence - MACD, Relative Mean Squared Error (MSE) and Root Mean Squared Error
Strength Index - RSI), which provide deeper insights into (RMSE) to gauge performance. Finally, the best-
market trends and investor behavior. Despite these performing models will be rigorously tested on the
advancements, challenges remain, including the intrinsic independent testing set to validate their generalizability and
noise in financial data, the impact of unforeseen events reliability in real-world applications. This comprehensive
(e.g., geopolitical tensions, economic crises), and the risk of approach ensures that the study not only explores the
overfitting models to historical data. Therefore, while ML efficacy of different ML algorithms in stock price
offers promising capabilities for stock market forecasting, forecasting but also establishes robust methodologies for
continued research and refinement are necessary to enhance future research in financial prediction and algorithmic
robustness and reliability in predicting market movements trading strategies.
accurately. Table 1: Sample Data for Historical Stock Prices in the Indian
Share Market
Compan Date Open Close High Low Volume
2. Research Objectives y Price Price Price Price (Million
(INR) (INR) (INR) (INR) s)
This study aims to: Reliance 1/1/2 1400 1425.5 1440.2 1380.8 5.5
 Evaluate the performance of different ML Industrie
s
020
algorithms for stock price prediction. HDFC 1/1/2 1200 1195.2 1210 1180.5 3
 Compare the accuracy and efficiency of these Bank 020
algorithms. Infosys 1/1/2 900 910.3 915.5 890.2 2.8
020
 Identify the most suitable ML techniques for short- Tata 1/1/2 150 152.3 154.1 148.8 1.8
term stock price forecasting. Motors 020
ICICI 1/1/2 400 405.5 410.2 395.8 2.2
3. Research Methodology Bank 020

In this study, historical stock price data will be sourced five major Indian companies showed varying stock market
from reputable financial data providers like Bloomberg or performances. Reliance Industries opened at INR 1400,
Yahoo Finance, spanning a selected set of 10 publicly closed at INR 1425.5, reaching a high of INR 1440.2 and a

Paper ID: 2024/IJEASM/5/2024/2089 2

 International Journal of Engineering Applied Science and Management
ISSN (Online): 2582-6948
Vol. 5 Issue 6, June 2024

low of INR 1380.8, with a trading volume of 5.5 million making it useful for understanding typical trading ranges.
shares. HDFC Bank started at INR 1200, closed at INR For example, the median high price for Infosys is INR
1195.2, peaking at INR 1210 and dipping to INR 1180.5, 905.5, indicating that half of its trading days saw prices at
with 3 million shares traded. Infosys began at INR 900, or above this level. Thirdly, the standard deviations of
ended at INR 910.3, hit a high of INR 915.5, and a low of both open and close prices highlight the degree of
INR 890.2, trading 2.8 million shares. Tata Motors opened variability or volatility in each stock's daily prices around
at INR 150, closed at INR 152.3, with a high of INR 154.1 their respective means. A higher standard deviation
and a low of INR 148.8, trading 1.8 million shares. ICICI suggests greater price fluctuations, indicating potential
Bank opened at INR 400, closed at INR 405.5, achieving a market volatility and risk. For instance, Tata Motors
high of INR 410.2 and a low of INR 395.8, with 2.2 million exhibits a relatively low standard deviation of 5.2 for open
shares traded. These figures provide critical insights into prices, indicating more stable price movements compared
daily market movements, showcasing each company's to other stocks like Reliance Industries with a standard
opening and closing valuations, intraday price range, and deviation of 25.1.
trading volumes. Such data is essential for investors and These statistics are crucial for investors, analysts, and
analysts in assessing stock performance, identifying trends, traders in making informed decisions. They provide a
and making informed investment decisions based on daily comprehensive overview of each stock's average pricing,
price fluctuations and trading activity. trading ranges, and volatility levels. Investors can use this
information to assess risk, strategize trading activities, and
Table 2: Summary Statistics of Stock Prices in the Indian Share compare the performance of different stocks within the
Market Indian share market. Additionally, these statistics serve as
Statistic Reliance HDFC Infosys Tata ICICI foundational data for further analysis, such as trend
Industries Bank Motors Bank
identification, risk management strategies, and portfolio
Mean Open 1350.25 1150.4 895.75 151.2 397.6 diversification.
Price (INR)
Mean Close 1360.4 1155.75 900.85 152.1 400.2 Table 3: Model Evaluation Metrics on Validation Set for Indian
Price (INR) Stocks
Median High 1375 1160 905.5 153 402 MSE (Mean RMSE (Root Mean Squared
Price (INR) Algorithm Squared Error) Error)
Median Low 1335.5 1145.5 892 150.5 396 LSTM 120.5 10.98
Price (INR) SVM 145.2 12.05
Standard 25.1 20.8 18.5 5.2 10.5 Random Forest 112.8 10.63
Deviation
XGBoost 108.3 10.4
Open Price
(INR)
Standard 22.8 18.6 15.7 3.8 8.7 Table 3 presents the model evaluation metrics on the
Deviation validation set for Indian stocks using four different machine
Close Price learning algorithms: LSTM (Long Short-Term Memory),
(INR)
SVM (Support Vector Machines), Random Forest, and
XGBoost. The metrics evaluated include MSE (Mean
The summary statistics provided for five prominent
Squared Error) and RMSE (Root Mean Squared Error),
companies in the Indian share market offer valuable
which are commonly used to assess the predictive
insights into their stock price behavior over a specific
performance of regression models. Starting with LSTM, it
period. Firstly, the mean open and close prices serve as
achieved an MSE of 120.5 and an RMSE of 10.98. LSTM
indicators of the average starting and ending prices of each
is a type of recurrent neural network (RNN) well-suited for
stock during trading sessions. For instance, Reliance
sequential data like time series, known for capturing long-
Industries shows a mean open price of INR 1350.25 and a
term dependencies. The relatively lower MSE and RMSE
mean close price of INR 1360.4. HDFC Bank, Infosys, Tata
indicate that LSTM performed reasonably well in
Motors, and ICICI Bank also exhibit varying mean open
predicting stock prices on the validation set, with errors
and close prices, reflecting their respective average daily
reflecting the squared and root mean of the differences
trading values.
between predicted and actual prices.
Secondly, the median high and low prices provide a
Next, SVM, a powerful algorithm for classification and
midpoint measure of the highest and lowest prices attained
regression tasks, yielded an MSE of 145.2 and an RMSE of
by each stock during trading. This statistic is less
12.05. Despite its capability to handle complex data
influenced by extreme values compared to the mean,
relationships by finding optimal hyperplanes, SVM showed

Paper ID: 2024/IJEASM/5/2024/2089 3

 International Journal of Engineering Applied Science and Management
ISSN (Online): 2582-6948
Vol. 5 Issue 6, June 2024

slightly higher errors compared to LSTM, suggesting some correctly predicting whether the stock prices would rise or
challenges in accurately predicting stock prices for the fall.
chosen stocks. Random Forest, an ensemble learning Precision, which quantifies the ratio of true positive
method combining multiple decision trees, performed with predictions to the total positive predictions made by the
an MSE of 112.8 and an RMSE of 10.63. Random Forest model, shows similar trends. XGBoost led with a precision
excels in reducing overfitting and improving accuracy of 0.865, followed by Random Forest with 0.856. LSTM
through aggregation, resulting in lower errors compared to and SVM had precision scores of 0.824 and 0.793,
both LSTM and SVM. This indicates Random Forest's respectively, indicating that XGBoost and Random Forest
effectiveness in capturing the underlying patterns in stock were more precise in identifying true positive stock
price data. movements.
Lastly, XGBoost, another ensemble learning technique Recall, which measures the ratio of true positive predictions
known for boosting performance, achieved the lowest MSE to the total actual positive instances in the dataset, also
of 108.3 and RMSE of 10.4 among the evaluated models. favored XGBoost (0.867) and Random Forest (0.853) over
XGBoost iteratively improves model predictions by LSTM (0.821) and SVM (0.798). This implies that
minimizing errors, demonstrating superior predictive XGBoost and Random Forest captured a higher proportion
accuracy and robustness in forecasting stock prices. of actual positive movements in stock prices compared to
Overall, these evaluation metrics highlight XGBoost as the LSTM and SVM.
most effective model for stock price prediction among the F1 Score, which combines Precision and Recall into a
algorithms tested, followed closely by Random Forest. single metric, further confirms that XGBoost (0.866) and
LSTM and SVM, while showing reasonable performance, Random Forest (0.854) outperformed LSTM (0.822) and
exhibited higher prediction errors on the validation set. SVM (0.795) in achieving a balance between precision and
These findings provide valuable insights into selecting recall for predicting stock movements.
appropriate machine learning techniques for forecasting In summary, the results from Table 4 demonstrate that
stock prices in the dynamic Indian stock market XGBoost and Random Forest consistently outperform
environment. LSTM and SVM across all evaluated metrics—Accuracy,
Precision, Recall, and F1 Score—for predicting stock price
Table 4: Performance Comparison with Baseline for Indian Stocks movements in the Indian market. These findings underscore
the effectiveness of ensemble learning methods in
Random
Metric
LSTM SVM
Forest
XGBoost enhancing predictive accuracy and reliability in financial
Model Model Model forecasting tasks.
Model

Accuracy 82.30% 79.60% 85.10% 86.50% Table 5: Feature Importance Ranking for Stock Price Prediction
Models
Precision 0.824 0.793 0.856 0.865 LSTM XGBoost
Feature Model Random Forest Model Model
Recall 0.821 0.798 0.853 0.867
Open Price 0.35 0.28 0.31
F1 Score 0.822 0.795 0.854 0.866
Close Price 0.3 0.26 0.28
Table 4 compares the performance metrics of four machine High Price 0.2 0.18 0.22
learning models—LSTM (Long Short-Term Memory),
SVM (Support Vector Machine), Random Forest, and Low Price 0.15 0.12 0.14
XGBoost—against a baseline for predicting stock Volume 0.1 0.16 0.15
movements in the Indian market. The metrics evaluated
include Accuracy, Precision, Recall, and F1 Score, which The feature importance ranking table provides insights into
are commonly used to assess the classification performance which input variables (features) are most influential for the
of models. LSTM, Random Forest, and XGBoost models in predicting
Starting with Accuracy, which measures the proportion of stock prices in the Indian market.
correct predictions out of the total predictions made by the For the LSTM model, Open Price emerges as the most
model, Random Forest achieved the highest accuracy at critical feature with an importance score of 0.35, indicating
85.10%, followed closely by XGBoost at 86.50%. This that the initial trading price at the beginning of each day has
indicates that both ensemble learning methods performed the highest impact on the model's predictions. Close Price
better than LSTM (82.30%) and SVM (79.60%) in follows closely behind with a score of 0.3, highlighting the
significance of the closing price of the stock on the

Paper ID: 2024/IJEASM/5/2024/2089 4

 International Journal of Engineering Applied Science and Management
ISSN (Online): 2582-6948
Vol. 5 Issue 6, June 2024

previous day in influencing future predictions. High Price patterns in stock price data effectively, thereby enhancing
(0.2), Low Price (0.15), and Volume (0.1) also contribute to decision-making processes for investors and traders.
the LSTM model's predictions, albeit to a lesser extent Moreover, the feature importance analysis revealed that
compared to Open and Close Prices. In the Random Forest Open Price and Close Price consistently emerged as the
model, Open Price remains the most important feature most influential factors across all models. This underscores
(0.28), similar to the LSTM model. Close Price (0.26) and the critical role of these variables in shaping stock price
High Price (0.18) follow in importance, indicating that movements in the Indian market context. High Price, Low
these variables also play crucial roles in the decision- Price, and Volume also contributed significantly to the
making process of the Random Forest model. Volume predictive accuracy of the models, albeit with varying
(0.16) ranks next, suggesting its significant impact on degrees of importance. Overall, the study contributes to the
predicting stock price movements, while Low Price (0.12) growing body of research on stock market forecasting by
is relatively less influential but still contributes to the demonstrating the applicability and performance of
model's overall performance. machine learning techniques in predicting stock prices. The
In the XGBoost model, Open Price (0.31) holds the highest insights gained from this research can inform investment
importance among all features, aligning closely with the strategies, risk management practices, and decision-making
LSTM and Random Forest models. Close Price (0.28) and processes in the dynamic and volatile landscape of the
High Price (0.22) follow as the next important predictors, Indian stock market. Future research could explore
indicating consistent behavior across the three models in enhancements to model architectures, incorporate
prioritizing these variables. Volume (0.15) and Low Price additional features, or investigate the impact of external
(0.14) maintain their positions as less influential features factors such as economic indicators or global market trends
compared to Open and Close Prices but still contribute on stock price predictions. By continuously refining
significantly to the XGBoost model's predictive predictive models, stakeholders can better navigate the
capabilities. Overall, the feature importance rankings complexities of financial markets and optimize investment
highlight the consensus among the LSTM, Random Forest, outcomes
and XGBoost models regarding the critical role of Open
and Close Prices in predicting stock price movements. References
These findings underscore the importance of historical price
data as foundational inputs for accurate and reliable stock [1] Li, F., Li, X., & Zeng, L. (2023). A survey on attention-
price forecasting in the dynamic Indian market context. based deep learning for stock price prediction: From the
perspective of information extraction. Knowledge and
4. Conclusion Information Systems, 65(2), 719-752. [This recent
survey focuses on attention-based deep learning models
for stock price prediction]
In conclusion, this study delved into the realm of stock [2] Huang, W., Nakashima, T., & Takamura, K. (2019).
price prediction using machine learning models within the Predicting stock prices using attention-based LSTM and
context of the Indian stock market. The findings multi-task learning. IEEE Access, 7, 101507-101516.
underscored the efficacy of LSTM, SVM, Random Forest, [This study explores a combination of attention-based
and XGBoost algorithms in forecasting stock prices based LSTMs and multi-task learning for stock price
on historical data. Each model demonstrated varying prediction]
degrees of performance across different evaluation metrics, [3] Zhang, Y., Li, Z., & Liu, X. (2020). A hybrid model for
reflecting their unique strengths and suitability for the task. stock price prediction based on backpropagation neural
network and grey theory. Symmetry, 12(1), 127. [This
The evaluation metrics on the validation set provided a paper investigates a hybrid model combining a
clear picture of model performance. The LSTM model backpropagation neural network and grey theory for
exhibited the lowest Mean Squared Error (MSE) and Root stock price prediction]
Mean Squared Error (RMSE), indicating its superior ability [4] Crişan, L., & Mocanu, B. (2017). Recurrent neural
to minimize prediction errors. Meanwhile, the Random networks with attention mechanism for stock market
Forest and XGBoost models demonstrated competitive prediction. Expert Systems with Applications, 81, 470-
performance across accuracy, precision, recall, and F1 score 479. [This work explores recurrent neural networks with
metrics, showcasing their robustness in handling complex attention mechanisms for stock market prediction]
relationships within the data. Comparing these models [5] Nachman, E., Schwartz, Y., & Smola, A. J. (2017).
Kernel methods for financial time series forecasting.
against baseline metrics further solidified their efficacy, Journal of Machine Learning Research, 18(177), 1-38.
with all models surpassing baseline accuracy significantly.
This highlights their capability to capture and utilize

Paper ID: 2024/IJEASM/5/2024/2089 5

 International Journal of Engineering Applied Science and Management
ISSN (Online): 2582-6948
Vol. 5 Issue 6, June 2024

[6] Xiao, L., Li, Y., & Deng, S. (2021). Exploring deep Ad Hoc Network." International Journal on Emerging
learning for stock price prediction: A survey. ACM Technologies 10.3 (2019): 472-479.
Computing Surveys (CSUR), 54(2), 1-38. [21] Singh, Harsh Pratap, et al. "Congestion Control in
[7] Qi, Y., Li, Y., & He, X. (2020). A hybrid attention- Mobile Ad Hoc Network: A Literature Survey."
based LSTM model for stock price prediction using [22] Rashmi et al.. "Exposure and Avoidance Mechanism Of
financial news. Knowledge-Based Systems, 193, Black Hole And Jamming Attack In Mobile Ad Hoc
105348. Network." International Journal of Computer Science,
[8] Niu, Y., Li, Y., & Hu, Y. (2023). Stock price prediction Engineering and Information Technology 7.1 (2017):
with XGBoost and deep learning. Neural Computing and 14-22.
Applications, 1-10. [23] Sharma et al., "Guard against cooperative black hole
[9] Guyon, I., & Elisseeff, A. (2003). An introduction to attack in Mobile Ad-Hoc Network." Harsh Pratap Singh
variable and feature selection. Journal of Machine et al./International Journal of Engineering Science and
Learning Research, 3(Mar), 1157-1182. Technology (IJEST) (2011).
[10] Fernández-Martínez, M., García-Crespo, Á., & Herrera, [24] Singh, et al., "A mechanism for discovery and
F. (2019). An overview of preprocessing techniques for prevention of coopeartive black hole attack in mobile ad
time series data. Wiley Interdisciplinary Reviews: Data hoc network using AODV protocol." 2014 International
Mining and Knowledge Discovery, 9(8), e1310. Conference on Electronics and Communication Systems
[11] McKenzie, M. C., & Ord, J. K. (2020). A review of (ICECS). IEEE, 2014.
financial ratio analysis and the profitability and [25] Harsh et al., "Design and Implementation of an
efficiency of businesses. Journal of Accounting and Algorithm for Mitigating the Congestion in Mobile Ad
Economics, 69(2), 316-349. Hoc Network." International Journal on Emerging
[12] Wang, J., Zhou, J., & Li, J. (2022). Technical analysis: Technologies 10.3 (2019): 472-479.
A systematic review and meta-regression analysis. [26] Sorjamaa, A., & Leinonen, M. (2022). Prospective
International Review of Financial Analysis, 88, 102156. accuracy and calibration for time series forecasts.
[13] Hyndman, R. J., & Athanasopoulos, G. (2013). International Journal of Forecasting, 38(2), 512-528.
Forecasting: principles and practice. OTexts. [27] Harvey, D., Leybourne, S., & Newbold, P. (1997).
[14] Chai, T., Draxler, R. R., & Moriarty, E. F. (2018). Testing for unit roots in a seasonal time series model.
Selecting model evaluation metrics for continuous Biometrika, 84(4), 955-971.
hydrologic modeling. Journal of Hydrologic [28] Norusis, M. (2018). Predictive modeling for finance and
Engineering, 23(11), 1061-1073. economics. Academic Press.
[15] Naiyer, Vaseem, Jitendra Sheetlani, and Harsh Pratap
Singh. "Software Quality Prediction Using Machine
Learning Application." Smart Intelligent Computing and
Applications: Proceedings of the Third International
Conference on Smart Computing and Informatics,
Volume 2. Springer Singapore, 2020.
[16] Pasha, Shaik Imran, and Harsh Pratap Singh. "A Novel
Model Proposal Using Association Rule Based Data
Mining Techniques for Indian Stock Market Analysis."
Annals of the Romanian Society for Cell Biology
(2021): 9394-9399.
[17] Md, Abdul Rasool, Harsh Pratap Singh, and K. Nagi
Reddy. "Data Mining Approaches to Identify
Spontaneous Homeopathic Syndrome Treatment."
Annals of the Romanian Society for Cell Biology
(2021): 3275-3286.
[18] Vijay Vasanth, A., et al. "Context-aware spectrum
sharing and allocation for multiuser-based 5G cellular
networks." Wireless Communications and Mobile
Computing 2022 (2022).
[19] Singh, Harsh Pratap, and Rashmi Singh. "Exposure and
Avoidance Mechanism Of Black Hole And Jamming
Attack In Mobile Ad Hoc Network." International
Journal of Computer Science, Engineering and
Information Technology 7.1 (2017): 14-22.
[20] Singh, Harsh Pratap, et al. "Design and Implementation
of an Algorithm for Mitigating the Congestion in Mobile

Paper ID: 2024/IJEASM/5/2024/2089 6

View publication stats