Semester 2

PROJECTS

Title: EXPLORATORY DATA ANALYSIS AND PREDICTIVE MODEL BUILDING OF MS ADMISSION DATASET USING MACHINE LEARNING: LOGISTIC REGRESSION, DECISION TREE, RANDOM FOREST, SUPPORT VECTOR MACHINE, NAÏVE BAYES CLASSIFIER  

Synopsis: As we are studying MS, we know the amount of pressure someone has to go through during MS admissions, there are different requirements for the admission process and after meeting them all one will get a chance into the MS course of a certain University and In this project, we have taken the dataset of graduate admissions and our goal is to find out some important features from this dataset through Explanatory Data Analysis (EDA) and then we applied machine learning classifiers on this data to observe which model best fits this data. We’ve applied five ML classifiers namely, Logistic Regression, Decision Tree, Random Forest, Support Vector (SVC), and Naive Bayes.

 Student Name :

-----------------------------------

Supervisor: Dr. Indrani Mukherjee 

Title: A comparative study of mobile brands on Flipkart via random forests 

Synopsis: This project aims to determine what factors, if any, are the most important metrics used by customers intending to make a purchase. To this end we will be using the random forest analysis technique. Specifically, we intend to use random forest classifier to aid us in determining which of the 13 factors under consideration carry the greatest weight for a customer intending to purchase a new mobile phone, provided that they have already decided upon a brand. This exercise can shed light onto what features phone manufacturers can focus upon in new and upcoming models in order to attract more customers. 

Student Name :

-----------------------------------

Supervisor: Dr. Sushovan Jana

Title:Diamonds Price Prediction Using Supervised Learning Methods 

Synopsis: The goal of the project is to build a model that can accurately predict the price of the diamond given its depth,color,clarity,table and its dimensions.This lesson will cover an analysis of diamonds. Some of the things that will be covered include the rich history of the diamond market and how to use the EDA techniques covered in this course to develop a quantitative understanding of it.The end goal is to build a predictive model of diamonds that is going to help us figure out whether a given diamond is a good deal or a rip-off.This lesson will give a better understanding of what goes into the price of a diamond.

Student Name :

-----------------------------------

Supervisor: Prof. Taranga Mukherjee

Title:BUSINESS STRATEGY MAKING BY MACHINE LEARNING 

Synopsis: Business strategy can be understood as the course of action or set of decisions which assist the entrepreneurs in achieving specific business objectives. It is nothing but a master plan that the management of a company implements to secure a competitive position in the market, carry on its operations, please customers and achieve the desired ends of the business. In business, it is the long-range sketch of the desired image, direction and destination of the organisation. It is a scheme of corporate intent and action, which is carefully planned and flexibly designed with the purpose of: • Directing efforts and behaviour • Gaining command over the situation. A business strategy is a combination of proactive actions on the part of management, for the purpose of enhancing the company’s market position and overall performance and reactions to unexpected developments and new market conditions. The maximum part of the company’s present strategy is a result of formerly initiated actions and business approaches, but when market conditions take an unanticipated turn, the company requires a strategic reaction to cope with contingencies. Hence, for unforeseen development, a part of the business strategy is formulated as a reasoned response.

Student Name :

-----------------------------------

Supervisor: Mr. Mayukh Bhattacharya 

Title:Stock Market prediction using LSTM and AI 

Synopsis:: We are predicting the High stock price of an organization. We have developed an application for predicting close stock price using LSTM algorithm. We have used datasets belonging to Amazon. In the future, we can extend this application for predicting cryptocurrency trading and also, we can add sentiment analysis for better predictions.

 In the future, the stock market prediction system can be further improved by utilizing a much bigger dataset than the one being utilized currently. This would help to increase the accuracy of our prediction models. Furthermore, other models of Machine Learning could also be studied to check for the accuracy rate resulted by them.We are predicting the High stock price of an organization. We have developed an application for

predicting close stock price using LSTM algorithm.


 Student Name :

-----------------------------------

Supervisor: Dr. Indrani Mukherjee 

Title:Loan approval prediction 

Synopsis:Banking processes use manual procedures to determine whether or not a borrower is suitable for a loan based on results. Manual procedures were most effective, but they were insufficient when there were a large number of loan applications. At that time, making a decision would take a long time. As a result, the loan prediction machine learning model can be used to assess a customer's loan status and build strategies.

This model extracts and introduces the essential features of a borrower that influence the customer's loan status. The prediction model not only helps the applicant but also helps the bank by minimizing the risk and reducing the number of defaulters. Finally, it produces the planned performance (loan status). These reports make a bank manager's job simpler and quicker.

This article discusses an approach toward predicting Loan Approval Status by the bank through utilizing basic elementary Data Science and Machine Learning techniques. 

Student Name :

-----------------------------------

Supervisor: Prof. Taranga Mukherjee

Title:Billionaires Dream 

Synopsis:Our thesis revolves around Forbes 2022 edition’s Billionaires list. There are a total of 2668 billionaires in this year’s list. For each of them, we look into 10 individual factors - rank, age, final worth, year of birth, category, source, country, city, country of citizenship and gender.

Based on these 10 factors, we are trying to predict whether these billionaires are self-made or not using 6 different supervised machine learning algorithms. We have taken the 10 factors as the 10 independent variables (x1,x2,…,x10), and the question whether these billionaires are self-made or not as the dependent variable (y). If they are self-made based on the necessary factors, we have denoted y as 1, if not, we have denoted y as 0.

We have performed our prediction in 6 different approaches. We have used 6 primary supervised machine learning algorithms – Logistic Regression, K Nearest Neighbour Classification (KNN), Naïve Bayes, Decision Trees, Random Forest, Support Vector Machine (SVM) along with Hyper-Parameter Tuning and Ensemble Techniques like Bagging.

Our main focus lies in being able to make our prediction whether the billionaires from this year’s (2022) Forbes billionaires list are self-made or not using the above-mentioned approaches, to engage in a comparative study on which algorithm performs the best, and to establish a significant relationship between the 10 factors that we have considered for each of the Billionaires with the question whether they are self-made or not. 

Student Name :

-----------------------------------

Supervisor:Prof. Taranga Mukherjee

Title:Predicting Model of Bank Term Deposit Subscription 

Synopsis:In this project we have built a classifier to predict whether or not a client will subscribe a term deposit. If the classifier has high accuracy, tha banks can arrange a better management of available resources by focusing on the potential customers “picked” by the classifier, which will improve their efficiency a lot. Besides, we plan to find out which factors are influential to customers’ decision, so that a more efficient and precise campaign strategy can be designed to help to reduce the costs and improve the profits. 

Student Name :

-----------------------------------

Supervisor: Prof. Taranga Mukherjee , Prof. Mayukh Bhattacharya