Mathematical Information Retrieval

Abstract

We present an overview of the NTCIR Math Tasks organized during NTCIR-10, 11, and 12. These tasks are primarily dedicated to techniques for searching mathematical content with formula expressions. In this chapter, we first summarize the task design and introduce test collections generated in the tasks. We also describe the features and main challenges of mathematical information retrieval systems and discuss future perspectives in the field.

Scientific Statement Classification over arXiv.org

“…Nevertheless, [21] is the first body of work we're aware of that broaches a "statement classification"-near task description for arXiv. More recently, a larger subset of arXMLiv has been used by the Math information retrieval (MathIR) community, who employ over a hundred thousand articles for benchmarking math-aware search systems [22].…”

“…In particular, the HTML dataset preserves the exact environment scoping of marked up statements; it allows us to create structured trees for mathematical expressions; and clearly and reliably separates away the styling from the content of the document. Our approach has been recognized by the MathIR community [22], who use the machine-readable formula representations for their investigations into formula retrieval.…”