Akshit Monga

Compiler Design for a custom programming language

Abstract: Implement a compiler in Java to convert the mini-custom programming language to Java byte code.

Perfect 11, Best squad predictor for soccer teams

Aim: To predict the best playing squad possible for a soccer team using data analytics.

Method : - Prepared ER diagram and schema for the possible data required for the project. Used historical data to create normalized player profiles. On the basis of normalized score of a player against a particular team, predicted whether to include that player in the squad or not.

Technologies used:- -> Html, CSS for the frontend. -> PHP for the backend. -> Oracle Database for storing records.

Find-Med - A medicine recommendation system

Aim and Objective :- * To recommend medicine to user based on their symptoms. * User would input their symptoms from which they are suffering and based on that appropriate medicine would be suggested. Programming Language Used :- Python, Algorithm used: Latent Dirichlet allocation (LDA) , K-means Clustering algorithm Platform : Linux. Abstract :- We designed a help dock under the field of BIOINFORMATICS which will suggest user about medicine (these recommendations will be actually prescribed from doctor under the same field), and make it easier for the user. We aimed to create an elasticsearch based medicine suggestion system. In order to make it easier for the user we presented a personal-ized, self-learning predictive text using Latent Dirichlet Allocation (LDA); to facilitate the search even more for the user, we used the client-server model using socket programming to add cross-platform compatability to the tool, over the a Wi-Fi network.

Checking app behaviour against app description

Aim and objective: To check whether the android app is behaving as advertised or not. To check whether an app can be malicious for user as if it is accessing user’s device without permission or if it is accessing those information that these kind of apps don’t require. Programming language used: Python and Java Algorithm used: Latent Dirichlet allocation (LDA) Platform : Linux. Abstract: * Fetched app description from Google playstore by method of crawling. * Implemented natural language processing to remove stopwords and filter language. * Implemented Latent Dirichlet allocation (LDA) algorithm for topic modelling. * Formed clusters of related apps and extracted android apps (apk) files. * Decrypted manifest files to extract permissions and sensitive api’s used by the app. * Detected the outliers i.e the apps that could be malicious for users

Data Compression using Huffman Tree

Abstract: Implemented Huffman encoding and decoding in C++ using binary tree for a given data.

Plagiarism Detection Software

Designed and developed a program for plagiarism detection using C language.

Compared the data from a test file and to files stored in a location on a local computer using three different algorithms and calculated percentage match. A percentage match above 30% was used to determine if the test file was original or not.

Algorithms used : KMP Algorithm, LCSS Algorithm and Rabin-Karp Algorithm. Key concepts used : String matching, degeneration property, longest proper prefix, use of hash function, hashing.

2048 game

Designed and developed a game known as 2048 in C language.

Objective of the game is to reach to the 2048 number tile by moving the tiles in left, right, top or down moments of number tiles. The tiles on merger grows in the exponents of two as moved.