Year: 2017
Type: dissertation
Abstract: We present Wiser, a new search engine for expert finding in academia. Our system is unsupervised and it jointly combines multiple classical language modeling techniques, based on text evidences, with Wikipedia knowledge, via entity linking. The expertise of each indexed expert is modeled by Wiser through a graph-based representation of Wikipedia entities and their relationships. Each expert-graph is further refined via proper computations (e.g. clustering and random walks) and eventually enhanced with the latent representation of entities learned with word embeddings. The effectiveness of our system is established over a large-scale experimental test over standard datasets which shows better performance than other state-of-the-art competitors published in top conferences, such as WWW 2016.
Author: CIFARIELLO, PAOLO
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6988100146