Graph-of-word and TW-IDF: new approach to ad hoc IR

By François Rousseau, Michalis Vazirgiannis. 2013

156 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Text Categorization as a Graph Classification Problem (2015, background): “, 1998; Mihalcea and Tarau, 2004) and more recently in ad hoc IR (Blanco and Lioma, 2012; Rousseau and Vazirgiannis, 2013).”
Context-Aware Sentence/Passage Term Importance Estimation For First Stage Retrieval (2019, methodology): “A rich research literature studies how to better estimate term importance (e.g., [2, 3, 5, 18, 27, 36]), however frequency-based signals continue to dominate in both industry and academia.”
SIFT Meets CNN: A Decade Survey of Instance Retrieval (2016, methodology): “ing. For example, Okapi BM25 [122] additionally takes into account the average document length and the current document length, as well as other hyper-parameters. The heuristic improved TW-IDF scheme [123] integrates a graph-based TF-like function. Both methods are variants of the local weighting sc”
N-gram IDF: A Global Term Weighting Scheme Based on Information Distance (2015, methodology): “Graph-of-word and TW-IDF: New Approach to Ad Hoc IR.”
Distance geometry and data science (2019, background): “ 9 we will consider a graph representation of sentences known as the graphof-words (Rousseau and Vazirgiannis 2013).”
Main Core Retention on Graph-of-Words for Single-Document Keyword Extraction (2015, methodology): “The edges were drawn between terms co-occurring within a fixed-size sliding window W of size 4 over the processed text, value consistently reported as working well [3,7,16,19,22].”
Graph-based term weighting for text categorization (2015, methodology): “That is, the higher the number of co-occurences of two terms in the document, the higher the weight of the corresponding edge (i.e., the weight of each edge will be equal to the number of co-occurences of its endpoints).”