Weideman, 2017. Responsible Design in Applied Linguistics: Theory and Practice

Peeters, Ward; Linnegar, John; Pretorius, Marilize; Vulovic, Marina

doi:10.1075/etc.10.1.10pee

Cited by 1 publication

(1 citation statement)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…33.1% of users clearly stated that it is becoming increasingly difficult to find target information. Users' dissatisfaction with wireless search is as high as 31.3%, and 39.7% of users indicate that they are only marginally satisfied with wireless search [6]. 39.5% of users do not use wireless search.…”

Section: Introductionmentioning

confidence: 99%

English Text Similarity Detection Algorithm Based on Animal Algorithm

2021

ZAP

View full text Add to dashboard Cite

The text similarity detection algorithm has a wide range of applications in the processing of massive natural language text information. Unlike simple and complete repetitive search, the complexity of natural language has caused great difficulties in the calculation of text semantic similarity. The Simhash algorithm does not involve the semantic information of the text, and cannot support the semantic problems of natural language processing such as synonyms and polysemes. Therefore, using the "dimensionality reduction" advantage of animal algorithms in English text processing and the efficiency of the retrieval process, aiming at its inability to recognize semantically similar text content, this paper studies the English text similarity detection algorithm of animal algorithms. Aiming at the shortcomings of simhash in the semantic similarity of text, this paper proposes a semantic code design based on the synonym word forest and context through the study of existing synonym expansion schemes. Based on the comprehensive improvement scheme, a semantic fingerprint generation algorithm incorporating synonym information is proposed, which solves the problem of similar texts that cannot identify replacement synonyms. The experiments in this paper show that after testing the sample data of the algorithm in this paper, under the condition of using k = 3 parameter determination, the accuracy and recall of correct identification are over 77%. In contrast, the two indexes of the traditional simhash algorithm and the word frequency statistical algorithm are only about 70%. This proves that the improved algorithm proposed in this paper has achieved relatively good results for identifying multiple similar modification situations, especially the problem of synonym replacement.

show abstract

Section: Introductionmentioning

confidence: 99%

English Text Similarity Detection Algorithm Based on Animal Algorithm

2021

ZAP

View full text Add to dashboard Cite

show abstract

Weideman, 2017. Responsible Design in Applied Linguistics: Theory and Practice

Cited by 1 publication

References 3 publications

English Text Similarity Detection Algorithm Based on Animal Algorithm

English Text Similarity Detection Algorithm Based on Animal Algorithm

Contact Info

Product

Resources

About