This research project aimed to develop a highly efficient and effective Chrome extension 
that could classify tweets containing hate speech, along with conducting sentiment and 
topic analysis. Hate speech is a persistent and concerning issue on Twitter, yet the platform 
has made little effort to address it. To address this challenge, this research performed a 
series of experiments, including the use of Support Vector Machine (SVM), Random 
Forest, and Long Short-Term Memory-based (LSTM) neural network classifiers. The 
results of the experiments showed that the SVM classifier, combined with word2vec [1] 
feature engineering, outperformed all other methods. 
Then developed the Chrome extension using a monolithic repository architecture, utilizing 
React and Django. By implementing this extension, users are able to automatically analyze 
live tweets and identify hate speech content, along with obtaining sentiment and topic 
analysis. The outcome of this research project could provide a significant contribution 
towards a more positive online environment and towards curbing the prevalence of hate 
speech on Twitter.

Can deep learning approaches achieve similar accuracy as statistical models in predicting 
the stock market? This research introduces an approach integrating Generative Adversarial 
Networks (GANs) with Long Short-Term Memory (LSTM) generators and Convolutional 
Neural Network (CNN) discriminators to predict stock trading actions. By analyzing and 
computing on historical Open, High, Low, Close and Volume (OHLCV) data of the stock 
market, we generate a wide range of input features using Fourier transforms, auto-encoders, 
and several technical indicators. Extreme Gradient Boosting (XGBoost) is employed to 
filter and extract important features. The extracted features are fed into the LSTM 
generator, whose output is then fed to the CNN discriminator to discriminate the buy, sell, 
or hold signals. Results demonstrate the model's effectiveness in capturing temporal and 
spatial patterns, with implications for refining stock trading strategies and enhancing 
financial decision-making for maximizing profits.

Hordeum vulgare, known as Barley in English and Jamara in Nepali, is used in the Dashain festival traditionally. For this project, the seeds of Barley were collected from the Kathmandu district of Nepal and observed in three different samples. Seeds sowed in this study were kept under different light conditions. The samples were kept under regular sunlight, LED light source with a color temperature of 6500K and in a dark room with no light source separately for a period of 15 days. If the plants were exposed to various kinds of light it was observed that growth of the plants is influenced by types of light. It was also found that each of the above-mentioned sources of light affected the plant growth differently. Out of the three samples that were taken; the first sample was exposed to direct sunlight while the second sample was exposed to light with a daylight color temperature of 6500 Kelvin and the third sample was placed in a room with no exposure to light. This experiment drew conclusions based on the differences in growth rate, the color of the leaves, the number of branches developed from a seed, the formation of roots, and the strength of anchorage.

Neural networks, inspired by the biological structure of the brain, have become a cornerstone in various machine learning tasks, including but not limited to those within the fashion domain. Convolutional Neural Networks (CNNs) are specialized neural networks designed for processing structured grid data - particularly images. Using these neural networks, The Virtual Wardrobe Fashion Advisor was created. Additionally programming languages like Python for algorithm implementation and Django,Html, Javascript and Cascading Style Sheets for web based application development were also used. The model was approached in the traditional CNN method to detect the type of clothes users prefer. Model detects the clothes with the help of selective gender and seasons; a setting in accordance to the user preference.

This project introduces SocketDB, a novel real-time SQL database system specifically engineered to meet the evolving demands of contemporary applications requiring instantaneous data updates and robust query processing capabilities. By integrating a WebSocket-based notification system, SocketDB offers a unique approach, allowing clients to subscribe and instantly receive updates, thus facilitating real-time interactions and analytics. Built on a columnar data layout, it ensures efficient query performance and storage optimization, particularly beneficial for read-intensive scenarios. The core of SocketDB is enhanced by a recursive descent SQL parser, enabling comprehensive support for a wide array of SQL operations, making it highly adaptable to complex data manipulation needs. Additionally, SocketDB incorporates advanced features such as data compression with the Zstandard algorithm, ensuring reduced storage footprint without compromising access speed. Embedded fault tolerance mechanisms and scalable infrastructure further reinforce its reliability and applicability in diverse environments. Aimed at developers and businesses alike, SocketDB is poised to transform real-time data handling, offering a scalable, efficient, and user-friendly platform for dynamic data management and analysis.