| Autor: |
Vedmiediev, D. O., Shapoval, N. V. |
| Předmět: |
|
| Zdroj: |
Electronics & Control Systems; 2023, Vol. 78 Issue 4, p16-20, 5p |
| Abstrakt: |
The division into groups of text messages is considered, which can be useful when building a personalized approach in different systems. Тo solve this problem, the Embedded Word2Vec was proposed. To enhance the division into groups, the suggestion of employing mini-batch k-means is presented, offering a method with lower computational demands. This recommendation aligns with the practical need for efficient and scalable clustering methods, especially when dealing with large datasets. Furthermore, the proposed metric based on the greatest common sequence is highlighted as a valuable tool for evaluating the similarity of texts. This metric not only serves as a means to assess clustering quality but also underscores the methodological approach of directly working with text data. The combination of these techniques presents a comprehensive framework for robust and effective text clustering, with potential applications in diverse fields, such as personalized system interactions and information retrieval. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
