Dr John Panneerselvam and Dr Angela Izah
Books2Door Ltd is a leading online bookseller, which offers exceptional books at affordable prices. With the motivation of promoting online book sales further, the company is looking to exploit the available sales trace logs, and other relevant dataset to become a data-driven and innovate e-commerce business.
This aim of this MPhil project is to develop analytics methodologies using heterogeneous data sources to generate predictive insights into customer buying behaviour, optimising pricing strategy, inventory management and to predict product performance. Successful applicant will conduct research to investigate state-of-the-art statistical, machine-learning and deep-learning techniques, as appropriate by addressing the contemporary issues in data management and analytics of heterogeneous sources of raw data. The outcome of this project will be the development of a novel data-driven decision support system for Books2Door, which can convert raw data into insights and present such insights in an intuitive dashboard for decision-making. The challenge and novelty lies in the development of an autonomous analytics framework encompassing the following features.
• Map the company metrics that will enable a better understanding of books that can be classed as profitable, high demand, low demand and less profitable, to facilitate right stocking of books for reducing cost, and generating more revenue and profit.
• Identify the optimal price point to maximise total profit, to deliver a competitive advantage within the pricing strategy
• Identify books that could attract a new market or expand current market share
• Understand the current customer base and their buying patterns to promote more sales to current customers
• Inventory slow moving goods, pinpoint these in inventory of 1 million books
• Forecasting price changes of competitors (based on analysing relevant datasets reflecting market trend)
• Predict bestselling books based on internet trends by analysing the correlation between the release of online web series and corresponding book sales.
• Forecasting general sales trends and demands in the market for appropriate book stacking
• Develop an intuitive dashboard to present the aforementioned insights for easy decision making