Rajdeep Gupta

-> Built a classification model on the dataset that consisted of 11959 instances, 56 continuous attributes and a nominal target Variable with 61 predictors and 2 classe. Used SMOTE and Resample filters.
-> Analyzed the dataset using different classification algorithms- Artificial neural networks, Support vector machines, Random forest classifier and Logistic regression.
-> TP and TN rates were evaluated by comparing models considering the performance to classify both classes… Show more

-> Built a classification model on the dataset that consisted of 11959 instances, 56 continuous attributes and a nominal target Variable with 61 predictors and 2 classe. Used SMOTE and Resample filters.
-> Analyzed the dataset using different classification algorithms- Artificial neural networks, Support vector machines, Random forest classifier and Logistic regression.
-> TP and TN rates were evaluated by comparing models considering the performance to classify both classes with good accuracy.
-> Built final model using Random forest classifier. Alterations were done with the parameters to achieve cross validation accuracy.
-> Prediction was calculated with 86% of accuracy. Show less

-> Developed a complete E-R model in an E-R diagram for Database for Course Timetabling.

-> Transformed the E-R model to a relational model and then implemented it in MS Access.

-> Incorporated the relational model in Access into a windows and web application using Visual Studio.

-> Added security and interface usability features to windows application.

->To give business model concept design providing it value proposition.

-> To give business model plan which includes revenue, breakeven point and profit.

-> To give information about customer segment, key activates performed by company.

-> To show growth of the company in future years and why our new product will succeed in market.

->Models were developed on the basis of product families and machine types using BInary coding to obtain Initial machine- product groups. Performed single pass heuristic by computing machine utilization.

->Used MVA tool to analyse the queue length , waiting time and other parameters for the various cells in the manufacturing plant.

->To find the most economically feasible model queuing and cost analysis was done to compare various models.

->Planned, Designed, conducted and analyzed a 2^3 full factorial experiment using Design of Experiments principles to find out the factors affecting comfortability of bike rider.

->Used replication, randomization and blocking techniques while conducting the experiment.

->Analyzed the results in JMP 12 Software package to find out the main effects and the two factor interaction effects. Model Adequacy checks were performed to investigate any violations in the assumptions.

->This project challenges the conventional braking system which aims to design and fabricate new type of master cylinder used in disc braking system.

->New design uses air vessel to trap air bubbles and hence improving braking efficiency by reducing turbulence, undesirable bleeding and vacuum generation when paddle is released. This would reduce the driver effort during the braking.
-> With this project we improved the braking efficiency by 4%.