Likhita Yakanuru

Dynamic and detail-oriented Data Analyst with over 1 years of experience in interpreting and analyzing data to drive growth for tech-centric organizations. Expert at designing and implementing effective data processing systems and models to mine through complex data sets, identify trends, and forecast outcomes. Proficient in utilizing advanced analytics tools and software to optimize data accuracy and integrity.

Intro to Nosql

Create a Data Project with Neo4j

Learning Ubuntu Desktop

Introduction to Linux

Apache Kafka Essential Training

MLOps Essentials: Model Development and Integration

Fine-tuned BERT for Rating Predictions:

A Sentiment Analysis Study on Amazon

In this project, I explore sentiment analysis on Amazon reviews using the fine-tuned BERT-base-cased model. Leveraging transfer learning, I trained the model to predict ratings from textual content, with Mean Squared Error (MSE) as the evaluation metric. The analysis includes assessing overall performance, examining predictions and attention mechanisms, and conducting error analysis to understand misclassifications. Additionally, I address trade-offs between model complexity and efficiency, optimizing for real-time applications. This research highlights the potential of state-of-the-art models in sentiment analysis, balancing performance and interpretability while addressing deployment challenges.

Software fault prediction based on Feature Selection Algorithm

Software Engineering is a field of computer science focused on enabling seamless communication between system software and user requirements. This project evaluates seven distinct machine learning algorithms using datasets from public promise repositories. The results aim to assist users in identifying and addressing defects while selecting the most efficient algorithm for their specific tasks, ensuring improved outcomes and overall effectiveness.