DE eng

Search in the Catalogues and Directories

Hits 1 – 20 of 20

1
Learning Embeddings for Rare Words Leveraging Internet Search Engine and Spatial Location Relationships ...
Abstract: Word embedding techniques depend heavily on the frequencies of words in the corpus, and are negatively impacted by failures in providing reliable representations for low-frequency words or unseen words during training. To address this problem, we propose an algorithm to learn embeddings for rare words based on an Internet search engine and the spatial location relationships. Our algorithm proceeds in two steps. We firstly retrieve webpages corresponding to the rare word through the search engine and parse the returned results to extract a set of most related words. We average the vectors of the related words as the initial vector of the rare word. Then, the location of the rare word in the vector space is iteratively fine-tuned according to the order of its relevances to the related words. Compared to other approaches, our algorithm can learn more accurate representations for a wider range of vocabulary. We evaluate our learned rare-word embeddings on the word relatedness task, and the experimental results ...
Keyword: Computational Algorithm; Computational Linguistics; Data Management System; FOS Languages and literature; Linguistics; Natural Language Processing; Semantics
URL: https://dx.doi.org/10.48448/jyvz-kx11
https://underline.io/lecture/29777-learning-embeddings-for-rare-words-leveraging-internet-search-engine-and-spatial-location-relationships
BASE
Hide details
2
Did the Cat Drink the Coffee? Challenging Transformers with Generalized Event Knowledge ...
BASE
Show details
3
Towards a Gold Standard Corpus for Variable Detection and Linking in Social Science Publications
In: Proceedings of the 11th International Conference on Language Resources and Evaluation (LREC) ; International Conference on Language Resources and Evaluation (LREC) ; 11 (2018)
BASE
Show details
4
Mining Social Science Publications for Survey Variables
In: Proceedings of the Second Workshop on NLP and Computational Social Science ; 47-52 (2018)
BASE
Show details
5
A Role And Reference account of interrogative sentences In Lakhota
In: The ITB Journal (2017)
BASE
Show details
6
Enhancing a Web Crawler with Arabic Search Capability
In: DTIC (2010)
BASE
Show details
7
Learning Structured Classifiers with Dual Coordinate Ascent
In: DTIC (2010)
BASE
Show details
8
THUIR at TREC 2009 Web Track: Finding Relevant and Diverse Results for Large Scale Web Search
In: DTIC (2009)
BASE
Show details
9
A Hybrid Approach for QA Track Definitional Questions
In: DTIC (2006)
BASE
Show details
10
Informedia at TRECVID 2003: Analyzing and Searching Broadcast News Video
In: DTIC (2004)
BASE
Show details
11
Prosody modeling with soft templates
2003
BASE
Show details
12
Progressive-Search Algorithms for Large-Vocabulary Speech Recognition
In: DTIC (1993)
BASE
Show details
13
Parsing the Voyager Domain Using Pearl
In: DTIC (1991)
BASE
Show details
14
The Inside/Outside Algorithm: Grammatical Inference Applied to Stochastic Context-Free Grammars
In: DTIC AND NTIS (1988)
BASE
Show details
15
On the Complexity of ID/LP Parsing
In: DTIC AND NTIS (1984)
BASE
Show details
16
An Algorithm for Parsing Flow Graphs
In: DTIC AND NTIS (1984)
BASE
Show details
17
Modeling Natural Language Information for Use in the Combination of Evidence Problem.
In: DTIC AND NTIS (1984)
BASE
Show details
18
A New Characterization of Attachment Preferences
In: DTIC (1983)
BASE
Show details
19
The Representation and Selection of Commonsense Knowledge for Natural Language Comprehensionn
In: DTIC AND NTIS (1976)
BASE
Show details
20
Automatic acquisition of synonyms from French UMLS for enhanced search of EHRs
In: http://natalia.grabar.perso.sfr.fr/publications/grabar-MIE2008.pdf
BASE
Show details

Catalogues
0
0
0
0
0
0
0
Bibliographies
0
0
0
0
0
0
0
0
0
Linked Open Data catalogues
0
Online resources
0
0
0
0
Open access documents
20
0
0
0
0
© 2013 - 2024 Lin|gu|is|tik | Imprint | Privacy Policy | Datenschutzeinstellungen ändern