YUGA PRADHA MP

- Extensive experience of 12 years in developing predictive systems and creating efficient algorithms to improve data quality, identifying evaluating, designing and implementing statisical analyses of gathered data to create analytic metrics and tools

- Skilled in designing ,building and deploying data analysis systems for large data sets
- Creating algorithms to extract information from large data sets
- Establishing efficient automated processes for model development,validation implementation and large-scale data analysis

Certified Scrum Master , Scrum Alliance

PGP Machine Learning and Artificial Intelligence - Feb' 2021

Predictive Modeling for Engagement:

• Developed predictive models using techniques such as logistic regression, random forests, and gradient boosting to forecast user engagement levels and predict churn risk with an accuracy of 85%, resulting in a 10% reduction in customer attrition over six months
• Leveraged historical user data to anticipate future engagement trends and proactively intervene to prevent attrition.
• Identify the target audience based on the demographic or behavioral characteristics of subscribers to focus on (e.g., past engagement patterns,purchase mode,trial users to subscribed users and so) and understand how increasing engagement can benefit the healthcare subscription service and plans.

Recommendation Systems:

• Designed and implemented a collaborative filtering recommendation system, increasing click-through rates on the company website by 15%.
• Successfully deployed a recommendation system within the subscription platform, providing personalized healthcare guidance tailored to users' purchase preferences.
• Achieved a 35% increase in user engagement, measured by interaction rates with recommended healthcare content and product offerings.
• Improved subscriber retention by 25% through the delivery of targeted recommendations aligned with users' past purchase behavior.
• Increased subscription plan adoption and retention rates through the delivery of value-added recommendations aligned with users' healthcare needs and purchase behaviors.

Website Page Visits Analysis for User Engagement Optimization

• Conducted in-depth analysis of page visit data to understand user behavior, identify patterns, and optimize website content and layout for enhanced user engagement.
• Identified high-traffic pages and popular content topics based on page visit frequency and duration.
• Implemented personalized content recommendations based on user browsing history and interests, resulting in a 15% increase in click-through rates.
• Utilized web analytics tools such as Google Analytics or Adobe Analytics to collect and analyze page visit data.

Sentiment Analysis:

• Conducted sentiment analysis on customer feedback using machine learning models such as VADER (Valence Aware Dictionary and Sentiment Reasoner) and supervised learning classifiers.
• Classified customer sentiments as positive, negative, or neutral to gauge overall customer satisfaction and identify areas of improvement.

Feedback Analysis and Iterative Improvement:

• Analyzed user feedback, surveys, and support tickets to identify pain points and areas for enhancement in the platform's user experience.
• Built text classification models to automatically categorize customer feedback into different issue categories such as billing, product quality, customer service, etc.
• Prioritized and escalated critical issues for prompt resolution, enabling proactive customer support and service enhancement.
• Collaborated with product development and design teams to implement iterative improvements and feature enhancements based on user feedback and data insights.

Performance Monitoring and Optimization:

• Established KPIs and performance metrics to track the effectiveness of segmentation-based marketing initiatives.
• Continuously monitored campaign performance and customer feedback to iterate and optimize marketing strategies for maximum impact.

Big Data Analytics:

• Utilized Apache Spark for processing large-scale datasets, leading to a 40% improvement in data processing speed compared to previous methods(like SSIS deployment on native machines)