Sucharitha Batchu

Biclustering algorithm is used to find the local patterns in the quantized historical data. The local patterns obtained are regarded as the trading rules for making decisions on the financial market.
• K Nearest Neighbor algorithm was implemented using Python by searching the K most similar instances.

We can detect the emergence of new topics on social media using links between users that are generated dynamically (intentionally or unintentionally) through replies, mentions, and retweets and these can be utilized for discovering hidden market needs.
• Developed using change point detection technique that uses SDNML coding.
• Implemented using python and MATLAB.

• Improved the efficiency of the marketing campaign of a Portuguese bank by analyzing client’s term deposit subscription decision
• Implemented using Tableau, Python, R. Worked on various machine learning classification algorithms to find the best fit for the project data

Identifies real-time fraudulent credit card transactions to avoid transactions unauthorized by the customers. Instantaneous fraud detection to stop transaction completion was built using PySpark and SparkSQL
• Designed a multi-stream processing system using AWS, Apache Kafka Lambda architecture and machine learning algorithms

• Developed a search engine to give out the relevant information based on user input using concepts of indexing, page ranking, term frequency, inverse document frequency and Rogue value approach
• Implemented in Tkinter using python, pickle data modules, performed recall and precision as performance metrics for classifying documents

A web-based application to ease the medical processes of detecting severe diseases and their treatment. Covering major human diseases like heart, kidney issues and cancer. This provides an easy, user-friendly and near-perfect prediction
• Built using R Shiny and HTML to integrate various diseases under one URL. applied eight different machine learning and deep learning algorithms(Linear, Bayesian, Random Forest, Extreme GB Boosting, Decision Trees, Genetic Algorithms and Neural Networks… Show more

A web-based application to ease the medical processes of detecting severe diseases and their treatment. Covering major human diseases like heart, kidney issues and cancer. This provides an easy, user-friendly and near-perfect prediction
• Built using R Shiny and HTML to integrate various diseases under one URL. applied eight different machine learning and deep learning algorithms(Linear, Bayesian, Random Forest, Extreme GB Boosting, Decision Trees, Genetic Algorithms and Neural Networks with Tensor Flow) and found that Extreme GB Boosting gives the most promising results in all the disease data sets.
Show less

Designed a marketing strategy that delivers profit that is four times the marketing costs. Used customer data to predict the churn probability and quantified the value of the customer as opposed to traditional clustering methods to allocate resources.
• Estimated customers to target by using of customer subscription data and advertisement budget and optimized by using linear, integer programming optimization
• Designed a dashboard using R Shiny to present trends and for classifying… Show more

Designed a marketing strategy that delivers profit that is four times the marketing costs. Used customer data to predict the churn probability and quantified the value of the customer as opposed to traditional clustering methods to allocate resources.
• Estimated customers to target by using of customer subscription data and advertisement budget and optimized by using linear, integer programming optimization
• Designed a dashboard using R Shiny to present trends and for classifying customers to target
• Calculated the significance of each customer and increased gains from retention using marketing plans
Show less