Freelance projects
I have also undertaken several engaging freelance projects and compelling case studies that demonstrate my passion for data analysis and problem-solving. Within this section, I will present a curated selection of these projects, showcasing my versatility and proficiency in applying data-driven techniques to real-world scenarios. These freelance endeavours include diverse undertakings, such as sentiment analysis utilizing advanced NLP techniques and contributing to customer churn prediction in the e-commerce sector. These projects highlight my proactive approach to delivering actionable insights and data-powered solutions, further complementing my professional journey and commitment to excellence.
Sentiment Analysis for Divorced Parents' Communication Enhancement
This sentiment analysis project was developed for a nonprofit organization dedicated to improving parenting dynamics for children of divorced parents. The objective of the project was to analyze messages exchanged between parents through their accounts on a website, using a sentiment classification system. The messages were labeled into three categories: aggressive, neutral, and positive, aimed at assessing the tone and emotional content of the communication.
The approach utilized in this project involved the application of TextBlob, a Python library for processing textual data. Leveraging supervised modeling techniques, the project aimed to predict and categorize messages based on their sentiment polarities. By employing natural language processing and machine learning, the model was trained to identify aggressive, neutral, and positive language patterns within the messages.
The significance of this sentiment analysis model lies in its potential to provide valuable insights into the emotional state of divorced parents during their communication. This information can assist the nonprofit organization in offering tailored support and guidance to parents, ultimately contributing to a more conducive co-parenting environment for the well-being of the children involved
Customer Churn Prediction
In this customer churn prediction project for an e-commerce store, I initiated the analysis by thorough data cleaning and null value removal. Subsequently, I performed descriptive statistics to gain valuable insights into the dataset. One key aspect of the analysis involved determining the churn frequency and creating visualizations based on factors such as gender, tenure, and other relevant subjects. To further enhance the predictive capabilities, I employed K-means cluster models, leveraging their advanced clustering techniques to effectively forecast customer churn.
