lahari kanagala

Motivated and detail-oriented aspiring Data Analyst with practical experience gained through a recent internship. Proficient in data collection, analysis, and interpretation, with a strong foundation in statistical methodologies and data visualization techniques. Skilled in using SQL for database querying and management, and experienced in creating insightful visualizations with Tableau. Passionate about leveraging Python for data analysis, automation, and exploring machine learning applications. Eager to apply analytical skills and knowledge in a dynamic environment to drive data-driven decision-making and continuous improvement.

• AWS certified Cloud Practitioner
• Certification of project Treasure Hunt Game using Python 3 completion
• Certificate of project Turtle Graphics initiation in java and python completion
• Certified for completion of Data Science Industrial program
• Project completion certificate for Hierarchical Clustering Linear Discriminant Analysis

• Ice Cream Ordering System

Many small businesses cannot afford to develop an application that could deliver their products to customers. So, this application will be helpful to such small businesses. Designed and implemented a user-friendly app that allows for customized ice cream selection with instant pricing information. Enabled tracking of the complete preparation status.

• Doc Bot

 Designed and implemented a sophisticated system where users can input their symptoms, receiving accurate disease predictions from the bot. Utilized the Natural Language Toolkit as a platform to identify symptoms reported by users. Enhanced user experience by displaying common co-occurring symptoms, allowing for input of relevant ones, and presenting a comprehensive symptom list. Implemented MLP algorithm to predict the most suitable disease based on symptoms.

• Two Level LSTM for Sentiment Analysis Using Lexicon Embedding and Polar Flipping

This is a two-level labeling strategy for sentiment texts. In the first stage, annotators are invited to label a large number of short texts with relatively pure sentiment orientations. Each sample is labeled by only one annotator. In the second stage, a relatively small number of text samples with mixed sentiment orientations are annotated, and each sample is labeled by multiple annotators. Here, we proposed a two level long short-term memory (LSTM) network to achieve two-level feature representation and classify the sentiment orientations of a text sample to utilize two labeled datasets.

• Weapon Detection Using YOLOV5

This project focuses on detection of weapons like pistol, rifle, gun, knife. Utilized YOLOV5 to enhance accuracy and efficiency in analyzing low quality and distinct images.