AjuMS

Data abundance in the e-commerce industry has been increasing since digitalization from 2010s. This data abundance is simply the user reviews in the comment section in a text format. So since the comment/review data is huge, nobody can sit and read all the comments to earn feedback about the particular product. Sentiment Analysis (SA) is so used for, determining the positive, neutral and negative comments based upon Natural Language Processing (NPL) If a customer/user post a comment like “This will be Apple’s best phone”, the “good” comment will be classified as positive comment and organization will improve the production in that domain. If a customer/user posts a comment like “Apple’s worst phone in the software segment”, the “worst” comment will be classified as a negative comment, and the organization will rectify the product quality. In this project, I aim to develop and build a Machine learning model using a "tweets" dataset that analyzes thousands of user comments (natural language) and provides a dedicated clustered result, so that a manager can understand whether the review overall is a good or a bad one.

This project is based on the problem statement: Business organizations face day-to-day competition around their organization. A business need is a need of a customer, so what and how customers like to buy next will be the target of every organization. This understanding can help us prevent oversupply, undersupply, and logistics costs. To find a customer's need, we need some data management, like big data. nowadays, businesses adopt ever-increasing precision marketing efforts to remain competitive and to maintain or grow their margin of profit. As such, forecasting models have been widely applied in precision marketing to understand and fulfill customer needs and expectations. Supply chain management (SCM) focuses on customer demands, flow of supply, entity relations, and goods management. Sometimes there will be demand uncertainties, organizational risks, supplies, transportation, and lead times, so that a data-driven approach is needed.