Mathematical Information Retrieval Trends and Techniques

Abstract

Mathematical formulas are widely used to express ideas and fundamental principles of science, technology, engineering, and mathematics. The rapidly growing research in science and engineering leads to a generation of a huge number of scientific documents which contain both textual as well as mathematical terms. In a scientific document, the sense of mathematical formulae is conveyed through the context and the symbolic structure which follows the strong domain specific conventions. In contrast to textual information, developed mathematical information retrieval systems have demonstrated the unique and elite indexing and matching approaches which are beneficial to the retrieval of formulae and scientific term. This chapter discusses the recent advancement in formula-based search engines, various formula representation styles and indexing techniques, benefits of formula-based search engines in various future applications like plagiarism detection, math recommendation system, etc.

Complementing Location-Based Social Network Data With Mobility Data: A Pattern-Based Approach

“…Then, we use standard information retrieval metrics to compare the produced rankings R L BS N and R tr for different k values. Specifically, P@k and R@k are commonly used to compare a ranked list with a reference one (namely, the ground truth) [31]. In our context, we apply them to verify the accuracy of trip sequences to model activities observed in LBSN data.…”