Ranking Experts in Location-aware Question-Answering Services

点击翻译按钮获取中文摘要

With the growing use of mobile devices, users increasingly expect to find more local and time-sensitiveinformation, such as the new popular restaurant or the hidden café around the corner. In this project, we aim toleverage big data and machine learning to address the problem of providing accurate and timely answers to userquestions from local experts. In particular, the industrial partner has developed a mobile app that facilitateslocation-based question-answering services. These services allow users to use their mobile phone to askquestions and find local information by interacting with local experts. A key to the success of these services isthe quality and timeliness of the responses provided by directing user questions to the right expert in adatabase. The main objective of this collaborative project is to address this problem by designing anddeveloping algorithms for finding and ranking experts in a database.The anticipated outcome of the research is twofold: (i) a framework for automatically enhancing a user'sexpertise profile, and (ii) novel algorithms, strategies and tools that can find and rank local experts that canprovide useful and timely answers. The scientific approach of our research is characterized by processing,modeling, analysis of large data sets and involves the use of advanced techniques for data analysis, such as textanalytics, machine learning, predictive analytics, and natural language processing.The proposed research collaboration aligns with Canada's Innovation Agenda. Conducting research in theintersection of big data analytics and machine learning has the potential to attract the brightest students fromaround the world, while keeping domestic talent here. Meanwhile, it will contribute to the training of highquality personnel, with first-rate analytical and problem-solving skills that have the capacity to stimulateknowledge-based companies and boost productivity and innovation taking place across Canada.