A Comprehensive Study of the Parameters in the Creation and Comparison of Feature Vectors in Distributional Semantic Models

Measuring the semantic similarity and relatedness of words is important for many natural language processing tasks. Although distributional semantic models designed for this task have many different parameters, such as vector similarity measures, weighting schemes and dimensionality reduction techniques, there is no truly comprehensive study simultaneously evaluating these parameters while also analysing the differences in the findings for multiple languages. We would like to address this gap with our systematic study by searching for the best configuration in the creation and comparison of feature vectors in distributional semantic models for English, Spanish and Hungarian separately, and then comparing our findings across these languages. During our extensive analysis we test a large number of possible settings for all parameters, with more than a thousand novel variants in case of some of them. As a result of this we were able to find such configurations that significantly outperform conventional configurations and achieve state-of-the-art results.

Download Full-text

Detecting linguistic idiosyncratic interests in autism using distributional semantic models

10.3115/v1/w14-3206 ◽

2014 ◽

Cited By ~ 4

Author(s):

Masoud Rouhizadeh ◽

Emily Prud'hommeaux ◽

Jan van Santen ◽

Richard Sproat

Keyword(s):

Semantic Models ◽

Distributional Semantic Models

Download Full-text

Professional Information Needs of Library Specialists in Russia

Bibliotekovedenie [Library and Information Science (Russia)] ◽

10.25281/0869-608x-2021-1-1-7-20 ◽

2021 ◽

Vol 70 (1) ◽

pp. 7-20

Author(s):

Marina Y. Neshcheret

Keyword(s):

Information Needs ◽

Rapid Development ◽

Information Service ◽

Library Science ◽

Information Process ◽

Cultural Workers ◽

Use Of Resources ◽

The Creation ◽

Professional Information ◽

Comprehensive Study

The article is devoted to the comprehensive study of the professional information needs (IN) of specialists of the central libraries of the subjects of the Russian Federation. The rapid development of high technologies in the field of accumulation, transmission and processing of information, the creation of modern telecommunications systems have led to the emergence of fundamentally new opportunities for organizing the information process. This, in turn, led to the qualitative growth of IN specialists, including those employed in the field of library activities. The specific features of IN library specialists are determined by their place and role in the modern process of cultural activity, industry orientation, nature of work and specialization. During 2018—2019, the Centre for Research of Problems of the Development of Libraries in the Information Society of the Russian State Library carried out research aimed at comprehensive study of the information needs of cultural workers employed in the library sphere. At the first stage of the research, there was comprehended the existing experience of studying professional information needs in the national library science. At the second stage, using the method of questionnaire survey, information needs of library specialists were studied in order to identify the most rational forms and methods of providing them. The analysis of the survey results made it possible to identify the sources of professional information, to reveal information resources that have the greatest importance and characterize specific features of librarians’ information needs.The author concludes on the need to expand the access to full-text databases and electronic versions of periodicals for library staff. The creation of integrated information centre could help providing library professionals with professional information. Currently, the function of such a centre is performed by the National Electronic Library, which includes the professional section for library specialists. The results of the study form theoretical and methodological basis for the rational use of resources and potential of libraries in providing information to professional information needs of library specialists and determine the prospects for further research related to improving the forms and methods of information service for this category of users.

Download Full-text

Semantic Memory Search and Retrieval in a Novel Cooperative Word Game: A Comparison of Associative and Distributional Semantic Models

Cognitive Science ◽

10.1111/cogs.13053 ◽

2021 ◽

Vol 45 (10) ◽

Author(s):

Abhilasha A. Kumar ◽

Mark Steyvers ◽

David A. Balota

Keyword(s):

Semantic Memory ◽

Memory Search ◽

Semantic Models ◽

Word Game ◽

Distributional Semantic Models ◽

Search And Retrieval

Download Full-text

Unsupervised Compositionality Prediction of Nominal Compounds

Computational Linguistics ◽

10.1162/coli_a_00341 ◽

2019 ◽

Vol 45 (1) ◽

pp. 1-57 ◽

Cited By ~ 3

Author(s):

Silvio Cordeiro ◽

Aline Villavicencio ◽

Marco Idiart ◽

Carlos Ramisch

Keyword(s):

Morphological Variation ◽

Red Wine ◽

Data Sets ◽

High Agreement ◽

Semantic Models ◽

Corpus Size ◽

Nominal Compounds ◽

Distributional Semantic Models ◽

The Impact

Nominal compounds such as red wine and nut case display a continuum of compositionality, with varying contributions from the components of the compound to its semantics. This article proposes a framework for compound compositionality prediction using distributional semantic models, evaluating to what extent they capture idiomaticity compared to human judgments. For evaluation, we introduce data sets containing human judgments in three languages: English, French, and Portuguese. The results obtained reveal a high agreement between the models and human predictions, suggesting that they are able to incorporate information about idiomaticity. We also present an in-depth evaluation of various factors that can affect prediction, such as model and corpus parameters and compositionality operations. General crosslingual analyses reveal the impact of morphological variation and corpus size in the ability of the model to predict compositionality, and of a uniform combination of the components for best results.

Download Full-text

Extraction of Terms Related to Named Rivers

Languages ◽

10.3390/languages4030046 ◽

2019 ◽

Vol 4 (3) ◽

pp. 46

Author(s):

Juan ◽

Faber

Keyword(s):

Knowledge Base ◽

Mississippi River ◽

Environmental Science ◽

Automatic Method ◽

Statistical Procedures ◽

Conceptual Structures ◽

Semantic Models ◽

Semantic Frames ◽

Conceptual Network ◽

Distributional Semantic Models

EcoLexicon is a terminological knowledge base on environmental science, whose design permits the geographic contextualization of data. For the geographic contextualization of landform concepts, this paper presents a semi-automatic method for extracting terms associated with named rivers (e.g., Mississippi River). Terms were extracted from a specialized corpus, where named rivers were automatically identified. Statistical procedures were applied for selecting both terms and rivers in distributional semantic models to construct the conceptual structures underlying the usage of named rivers. The rivers sharing associated terms were also clustered and represented in the same conceptual network. The results showed that the method successfully described the semantic frames of named rivers with explanatory adequacy, according to the premises of Frame-Based Terminology.

Download Full-text