Rani Godase

Proactive alert management – leveraged multiple sources of data to proactively manage the alert situations. Built POC by analyzing data of different forms and from different sources, found out correlation between them using statistical methods. Forecasting and capacity planning - Forecasted the volumes accurately using time series techniques to arrive at optimal staffing for improving the customer experience. Analyzed and understand the business requirement to translate into conceptual and logical data models. Collecting, collating and carrying out complex data analysis in support customer requests. Created detailed reports by collating all the reports for further analysis. Analyzed data from different sources and arranged them in structured format. Analyzed data in python and MS Excel.

Classification of Myers-Briggs Type Personality Indicator using Unimodal and Bi-modal Approaches (Research project). Objective: This thesis aims to research if there are any useful and predictable features in the data which lead to MBTI personality classification using deep learning techniques with facial images and social media posts. Found out useful and predictable features in the data which lead to personality classification. Discovered the use of a bi-modal approach, that is the combination of text and image features, leads to better classification than the uni-modal approach. Experimented with various machine learning and deep learning models to understand which features provide the most valuable information which leads to better prediction. Tools & Techniques: FaceNet, Natural Language Processing (NLP), CNN, Transfer Learning, VGG Face Models, SVM. Optimize the results obtained from ML technique using optimization algorithm. Applied suitable technique from research paper on given objective.