Logo des Repositoriums
 
Textdokument

Word Embeddings for Practical Information Retrieval

Lade...
Vorschaubild

Volltext URI

Dokumententyp

Zusatzinformation

Datum

2017

Zeitschriftentitel

ISSN der Zeitschrift

Bandtitel

Verlag

Gesellschaft für Informatik, Bonn

Zusammenfassung

We assess the suitability of word embeddings for practical information retrieval scenarios. Thus, we assume that users issue ad-hoc short queries where we return the first twenty retrieved documents after applying a boolean matching operation between the query and the documents. We compare the performance of several techniques that leverage word embeddings in the retrieval models to compute the similarity between the query and the documents, namely word centroid similarity, paragraph vectors, Word Mover’s distance, as well as our novel inverse document frequency (IDF) re-weighted word centroid similarity. We evaluate the performance using the ranking metrics mean average precision, mean reciprocal rank, and normalized discounted cumulative gain. Additionally, we inspect the retrieval models’ sensitivity to document length by using either only the title or the full-text of the documents for the retrieval task. We conclude that word centroid similarity is the best competitor to state-of-the-art retrieval models. It can be further improved by re-weighting the word frequencies with IDF before aggregating the respective word vectors of the embedding. The proposed cosine similarity of IDF re-weighted word vectors is competitive to the TF-IDF baseline and even outperforms it in case of the news domain with a relative percentage of 15%.

Beschreibung

Galke, Lukas; Saleh, Ahmed; Scherp, Ansgar (2017): Word Embeddings for Practical Information Retrieval. INFORMATIK 2017. DOI: 10.18420/in2017_215. Gesellschaft für Informatik, Bonn. PISSN: 1617-5468. ISBN: 978-3-88579-669-5. pp. 2155-2167. Deep Learning in heterogenen Datenbeständen. Chemnitz. 25.-29. September 2017

Zitierform

Tags