Unsupervised Does Not Mean Uninterpretable: The Case for Word Sense
            Induction and Disambiguation

Panchenko, Alexander; Ruppert, Eugen; Faralli, Stefano; Ponzetto, Simone Paolo; Biemann, Chris

doi:10.18653/v1/e17-1009

Cited by 26 publications

(22 citation statements)

References 36 publications

(59 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These induced senses do not necessarily correspond to human notions of sense distinctions, or are not easily distinguishable. For this reason, methods have been proposed to improve the interpretability of unsupervised sense representations, either by extracting their hypernyms or their visual representations (i.e., an image illustrating a specific meaning) (Panchenko et al, 2017b) or by mapping the induced senses to external sense inventories (Panchenko, 2016).…”

Section: Interpretabilitymentioning

confidence: 99%

From Word To Sense Embeddings: A Survey on Vector Representations of Meaning

Camacho-Collados¹,

Pilehvar²

2018

jair

230

View full text Add to dashboard Cite

Over the past years, distributed semantic representations have proved to be effective and flexible keepers of prior knowledge to be integrated into downstream applications. This survey focuses on the representation of meaning. We start from the theoretical background behind word vector space models and highlight one of their major limitations: the meaning conflation deficiency, which arises from representing a word with all its possible meanings as a single vector. Then, we explain how this deficiency can be addressed through a transition from the word level to the more fine-grained level of word senses (in its broader acceptation) as a method for modelling unambiguous lexical meaning. We present a comprehensive overview of the wide range of techniques in the two main branches of sense representation, i.e., unsupervised and knowledge-based. Finally, this survey covers the main evaluation procedures and applications for this type of representation, and provides an analysis of four of its important aspects: interpretability, sense granularity, adaptability to different domains and compositionality.

show abstract

Section: Interpretabilitymentioning

confidence: 99%

From Word To Sense Embeddings: A Survey on Vector Representations of Meaning

Camacho-Collados¹,

Pilehvar²

2018

jair

230

View full text Add to dashboard Cite

show abstract

“…Babelfy is based on a dense subgraph heuristic for selecting coherent semantic interpretations of the input text. Panchenko et al [49] proposed an unsupervised and knowledge-free word sense induction and disambiguation approach that relies on induced inventories as a pivot for learning sense feature representations.…”

Section: Related Workmentioning

confidence: 99%

ShotgunWSD 2.0: An Improved Algorithm for Global Word Sense Disambiguation

Butnaru

Ionescu

2019

IEEE Access

View full text Add to dashboard Cite

ShotgunWSD is a recent unsupervised and knowledge-based algorithm for global word sense disambiguation (WSD). The algorithm is inspired by the Shotgun sequencing technique, which is a broadly-used whole genome sequencing approach. ShotgunWSD performs WSD at the document level based on three phases. The first phase consists of applying a brute-force WSD algorithm on short context windows selected from the document, in order to generate a short list of likely sense configurations for each window. The second phase consists of assembling the local sense configurations into longer composite configurations by prefix and suffix matching. In the third phase, the resulting configurations are ranked by their length, and the sense of each word is chosen based on a majority voting scheme that considers only the top configurations in which the respective word appears. In this paper, we present an improved version (2.0) of ShotgunWSD which is based on a different approach for computing the relatedness score between two word senses, a step that stays at the core of building better local sense configurations. For each sense, we collect all the words from the corresponding WordNet synset, gloss and related synsets, into a sense bag. We embed the collected words from all the sense bags in the entire document into a vector space using a common word embedding framework. The word vectors are then clustered using k-means to form clusters of semantically related words. At this stage, we consider that clusters with fewer samples (with respect to a given threshold) represent outliers and we eliminate these clusters altogether. Words from the eliminated clusters are also removed from each and every sense bag. Finally, we compute the median of all the remaining word embeddings in a given sense bag to obtain a sense embedding for the corresponding word sense. We compare the improved ShotgunWSD algorithm (version 2.0) with its previous version (1.0) as well as several state-of-the-art unsupervised WSD algorithms on six benchmarks: SemEval 2007, Senseval-2, Senseval-3, SemEval 2013, SemEval 2015, and overall (unified). We demonstrate that ShotgunWSD 2.0 yields better performance than ShotgunWSD 1.0 and some other recent unsupervised or knowledge-based approaches. We also performed paired McNemar's significance tests, showing that the improvements of ShotgunWSD 2.0 over ShotgunWSD 1.0 are in most cases statistically significant, with a confidence interval of 0.01. INDEX TERMS Word sense disambiguation, shotgun sequencing, word embeddings, outlier removal.

show abstract

“…(3) a combination of cosine and Jaro-Winkler [43] measures is used to calculate the similarity score as follows.…”

Section: Word Sense Disambiguation Of the Conceptsmentioning

confidence: 99%

“…Four criteria are considered for calculating the contribution score: (1) lexical cohesion, (2) co-occurrences cohesion, (3) semantic cohesion and (4) Hierarchical path proximity. Lexical Cohesion: calculates the lexical cohesion of each concept in a message with all the remaining concepts in the same message by measuring Jaro-Winkler [43] similarity between them. This will give us the idea of how much each lexical entity is compatible with the information-content of message.…”

Section: 1 Contribution Score For Semantic Networkmentioning

confidence: 99%

Automatic Hashtag Recommendation in Social Networking and Microblogging Platforms Using a Knowledge-Intensive Content-based Approach

Jaderyan

Khotanlou

2019

IJE

View full text Add to dashboard Cite

A B S T R A C TIn social networking/microblogging environments, #tag is often used for categorizing messages and marking their key points. Also, since some social networks such as twitter apply restrictions on the number of characters in messages, #tags can serve as a useful tool for helping users express their messages. In this paper, a new knowledge-intensive content-based #tag recommendation system is introduced. The proposed system works by integrating structured knowledge in every core component. First, the relevant features, semantic structures and information-content are extracted from messages. Since little information can often be placed in a message, a content enrichment module is introduced to identify information structures that can improve the representation of message. The extracted features are represented by semantic network. Then, a hybrid and multi-layered similarity module identifies the commonalities and differences of the features, semantics and information-content in messages. At the end, #tags are recommended to users based on #tags in contextually similar messages. The system is evaluated on Tweets2011 dataset. The results suggests that the proposed method can recommend suitable #tags in negligible operational time and when little content is available.

show abstract

Unsupervised Does Not Mean Uninterpretable: The Case for Word Sense Induction and Disambiguation

Cited by 26 publications

References 36 publications

From Word To Sense Embeddings: A Survey on Vector Representations of Meaning

From Word To Sense Embeddings: A Survey on Vector Representations of Meaning

ShotgunWSD 2.0: An Improved Algorithm for Global Word Sense Disambiguation

Automatic Hashtag Recommendation in Social Networking and Microblogging Platforms Using a Knowledge-Intensive Content-based Approach

Contact Info

Product

Resources

About