Predictive Business Analytics for Incomplete Data

点击翻译按钮获取中文摘要

Business analytics refers to the process of continuous iterative exploration and investigation of relevant data to improve a company's performance. Business analytics relies on business insights and expert knowledge, analytics methodologies and technologies, as well as a huge amount of relevant data. A common type of business analytics is predictive analytics which involves exploiting business data to make predictions. Popular applications of predictive analytics include stock market prediction, customer lead generation, marketing campaign prediction, product bundling. The proposed project will focus on the theories and applications of predictive analytics. Currently, predictive analytics typically employs traditional statistics, data mining, artificial intelligence and machine learning techniques for analyzing data. Many of these techniques generally require that there is no missing values in the data. However, real-life business data often contain missing values. For example, when data are collected from customers directly, many customers may be unwilling to provide all information because of privacy concerns. In the research literature, data with missing values are referred to as "incomplete data". The primary objective of this project is to develop new predictive analytics methods for incomplete data. These proposed methods will be studied and validated through various business applications. Results from this project will significantly contribute toward the theories and applications of business analytics. We expect that the research results will be presented at international conferences and published in journals in the fields of computer science and business administration.