Bridging the theoretical gap between semantic representation models without the pressure of a ranking: some lessons learnt from LSA

Jorge-Botana, Guillermo; Olmos, Ricardo; Luzón, José María

doi:10.1007/s10339-019-00934-x

Cited by 17 publications

(11 citation statements)

References 114 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These models formally represent the semantic meaning of words as coordinates in a high‐dimensional vector space. In general, vector space models provide high‐quality input to emulate different cognitive mechanisms (see Günther et al., 2019; Jones, Gruenenfelder, & Recchia, 2018; or Jorge‐Botana, Olmos, & Luzón, 2020 for a recent revision on the use of these models to simulate cognitive process). Given that the semantic information of vector space models is derived from co‐occurrences of written words in texts (usually from a corpus of tens of thousands of documents), their semantic representations are considered amodal.…”

Section: Introductionmentioning

confidence: 99%

“…The methodology that uses the representations of LSA to generate different vector spaces representing different stages of vocabulary development is known as Word Maturity (Kireyev & Landauer, 2011; Landauer et al., 2011). LSA is well suited for this methodology because all words and texts are expressed on an orthogonal basis and this makes it possible to keep the similarity distances even with the transformations performed by the procedure (see Jorge‐Botana et al., 2020 for a discussion of the advantages of orthogonality). A brief explanation on how Word Maturity works will be given now.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Emotional Valence Precedes Semantic Maturation of Words: A Longitudinal Computational Study of Early Verbal Emotional Anchoring

2021

Self Cite

View full text Add to dashboard Cite

We present a longitudinal computational study on the connection between emotional and amodal word representations from a developmental perspective. In this study, children's and adult word representations were generated using the latent semantic analysis (LSA) vector space model and Word Maturity methodology. Some children's word representations were used to set a mapping function between amodal and emotional word representations with a neural network model using ratings from 9-year-old children. The neural network was trained and validated in the child semantic space. Then, the resulting neural network was tested with adult word representations using ratings from an adult data set. Samples of 1210 and 5315 words were used in the child and the adult semantic spaces, respectively. Results suggested that the emotional valence of words can be predicted from amodal vector representations even at the child stage, and accurate emotional propagation was found in the adult word vector representations. In this way, different propagative processes were observed in the adult semantic space. These findings highlight a potential mechanism for early verbal emotional anchoring. Moreover, different multiple linear regression and mixed-effect models revealed moderation effects for the performance of the longitudinal computational model. First, words with early maturation and subsequent semantic definition promoted emotional propagation. Second, an interaction effect between age of acquisition and abstractness was found to explain model performance. The theoretical and methodological implications are discussed.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Emotional Valence Precedes Semantic Maturation of Words: A Longitudinal Computational Study of Early Verbal Emotional Anchoring

2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…To build the topic model of LSA, "stop words" such as "the," "is," and "at" must be removed [18]. ese stop words have little substantive meaning for the topic model.…”

Section: Design Of Data Processing Modulementioning

confidence: 99%

Automatic Scoring of Spoken Language Based on Basic Deep Learning

Zhong

Zong-hua

2022

Scientific Programming

View full text Add to dashboard Cite

The oral English test in domestic universities requires teachers to modify a large number of candidates’ oral recordings. This is the work of using time repeatedly. Using the CALL system to realize the automation of conversation recording can reduce the burden of teachers’ work. Therefore, it is of great practical significance to develop an automatic and accurate scoring system for oral English. With the development of artificial intelligence, deep learning technology has been gradually applied in various fields. Similarly, in the application of oral scoring, deep learning technology makes the implementation of such a system possible. Based on the deep learning technology, this paper proposes an automatic scoring algorithm for spoken language and implements a detailed design and evaluation system. The system consists of two modules. The pronunciation standard of spoken pronunciation and the content of spoken pronunciation are scored, and the sum of these two scores is the final score. Finally, this paper uses 650 oral English recordings from a college English test to train the artificial neural network. Experimental results show that if the training data set is small, the BP network model can obtain better comprehensive evaluation performance.

show abstract

“…Vector techniques are based on the automatic processing of large text corpora representing a language of a general or specific domain, although some studies recommend specific domain corpora (Kwantes et al, 2016). There are different computational models in which word occurrences are algebraically vectorized such as LSA, word2vec, or BEAGLE (for a revision on space models see Günther, Rinaldi, & Marelli, 2019;Jorge-Botana, Olmos, & Luzón, 2020;Jones, Willits, & Dennis, 2015;or McNamara, 2011). All of them coincide in that they represent the lexicon in a reduced dimensionality vector space.…”

Section: Introductionmentioning

confidence: 99%

Can personality traits be measured analyzing written language? A meta-analytic study on computational methods

Moreno

Martínez-Huertas

Olmos

et al. 2021

Personality and Individual Differences

Self Cite

View full text Add to dashboard Cite

Bridging the theoretical gap between semantic representation models without the pressure of a ranking: some lessons learnt from LSA

Cited by 17 publications

References 114 publications

Emotional Valence Precedes Semantic Maturation of Words: A Longitudinal Computational Study of Early Verbal Emotional Anchoring

Emotional Valence Precedes Semantic Maturation of Words: A Longitudinal Computational Study of Early Verbal Emotional Anchoring

Automatic Scoring of Spoken Language Based on Basic Deep Learning

Can personality traits be measured analyzing written language? A meta-analytic study on computational methods

Contact Info

Product

Resources

About