Mining Market Trend from Blog Titles Based on Lexical Semantic Similarity

2012 ◽  
pp. 261-273 ◽  
Author(s):  
Fei Wang ◽  
Yunfang Wu
Keyword(s):  
Semantic Similarity ◽  
Lexical Semantic

The verbs ‘rain’ and ‘snow’ in Gallo-Romance, and other morphological mismatches in diachrony

2020 ◽  
pp. 385-402
Author(s):  
Martin Maiden
Keyword(s):  
Semantic Similarity ◽  
Morphological Analysis ◽  
Morphological Changes ◽  
Prima Facie ◽  
Lexical Meaning ◽  
Lexical Semantic ◽  
Word Forms ◽  
Entire Word

The historical morphology of the verb ‘snow’ in Francoprovençal presents a conundrum, in that it is clearly analogically influenced by the verb ‘rain’, for obvious reasons of lexical semantic similarity, but the locus of that influence is not the ‘root’ (the ostensible bearer of lexical meaning) but desinential inflexion-class members, which are in principle independent of any lexical meaning. Similar morphological changes are also identified for other Gallo-Romance verbs. It seems, in effect, that speakers can identify exponents of the lexical meaning of word-forms in linear sequences larger than the apparent ‘morphemic’ composition of those word-forms, even when such a composition may seem prima facie transparent and obvious. It is argued that these facts are inherently incompatible with ‘constructivist’, morpheme-based, models of morphology, and strongly compatible with what have been called ‘abstractivist’ (‘word-and-paradigm’) approaches, which generally take entire word-forms as the primary units of morphological analysis.

Unsupervised measure of Chinese lexical semantic similarity using correlated graph model for news story segmentation

2018 ◽  
Vol 318 ◽  
pp. 236-247 ◽  
Author(s):  
Wei Feng ◽  
Xuecheng Nie ◽  
Yujun Zhang ◽  
Lei Xie ◽  
Jianwu Dang
Keyword(s):  
Semantic Similarity ◽  
Graph Model ◽  
News Story ◽  
Lexical Semantic ◽  
Story Segmentation ◽  
News Story Segmentation

scholarly journals Developing a Semantic Similarity Judgment Test for Persian Action Verbs and Non-action Nouns in Patients With Brain Injury and Determining its Content Validity

2020 ◽  
Vol 21 (2) ◽  
pp. 154-181
Author(s):  
Tabassom Azimi ◽  
◽  
Zahra-Saddat Qoreishi ◽  
Reza Nilipour ◽  
Morteza Farazi ◽  
...  
Keyword(s):  
Semantic Similarity ◽  
Content Validity ◽  
Semantic Processing ◽  
Similarity Judgment ◽  
Point Scale ◽  
Action Verb ◽  
Lexical Semantic ◽  
Verb Processing ◽  
Action Verbs ◽  
The Brain

Objective: Brain trauma evidences suggest that the two grammatical categories of noun and verb are processed in different regions of the brain due to differences in the complexity of grammatical and semantic information processing. Studies have shown that the verbs belonging to different semantic categories lead to neural activity in different areas of the brain, and action verb processing is related to the activity of motor and pre-motor areas of the brain. Researchers use different tasks to evaluate action verb processing. The most common tasks are action naming and action fluency tasks. Although these types of tasks are sensitive to deficits in action verb processing, they do not specify the nature of the injury. To understand whether dysfunction in action verb processing is due to difficulty in lexical access or specific impairment in semantic processing, it is necessary to design a specific test to evaluate lexical-semantic processing. Semantic Similarity Judgment (SSJ) test targets the lexical-semantic encoding at a deep and controlled processing level. The purpose of the present study was to develop a SSJ test for Persian action verbs and non-action nouns and determine its content validity. Materials & Methods: In this methodological study, 70 Persian action concrete verbs and 80 Persian non-action concrete nouns were first selected. For each word, a semantically related word based on functional, physical, categorical features and similarity in action was selected according to the opinion of 4 experts (3 speech-language pathologists and one linguist) using a 7-point scale. For semantic similarity rating, only the pairs of words with a high semantic similarity score (5 to 7) remained and the rest were omitted. Then, for each pair of semantically related words, a semantically unrelated word was selected. After determining content validity qualitatively by three experts and removing inappropriate items, for matching the two sets of nouns and verbs, the lexical and psycholinguistic characteristics of the remaining words (207 nouns and 156 verbs) including frequency, number of syllables, phonemes, letters, phonological and orthographic neighbors, action association, imageability, familiarity and age of acquisition were extracted by 18 volunteers (13 speech-language pathologists and linguists and 5 parents selected by a convenience sampling method) based on a 7-point scale. The verbs with low action associations and the nouns with high action association were removed and then, the two sets of words were matched for other lexical and psycholinguistic characteristics. Finally, 34 triples of verbs with high action association and 34 triples of nouns with low action association were selected. In both noun and verb sets, the words were chosen in such a way that, in order to judge, the semantic features of the words need to be carefully considered. Data analysis was performed using descriptive statistics and independent t-test.

An Improved Tendentious Algorithm of Lexical Semantic

2014 ◽  
Vol 513-517 ◽  
pp. 1326-1331
Author(s):  
Yu Bing Yang ◽  
Yan Yan Wu ◽  
Zhan Ping Li ◽  
Lin Ge Wang
Keyword(s):  
Semantic Similarity ◽  
Practical Application ◽  
Accuracy Rate ◽  
Lexical Semantic ◽  
Experiment Validation ◽  
Similarity Calculation

Lexical semantic tendentious recognition calculation is the base of the sentence, and the sentence tendentious recognition is the text and the chapter structure tendentious recognition foundation. Based on HowNet lexical semantic similarity calculation, according to the current vocabulary appraisable tendentious theory, the paper put forward an improved orientation algorithm. With experiment validation, in the same pair of benchmark words, accuracy rate has lager improve to above 90%. There are some values in practical application.

Sign in / Sign up

Export Citation Format

Share Document