| Literature DB >> 36133160 |
Bo Li1,2.
Abstract
Primary and secondary schools have the enormous responsibility of developing talent over a hundred years, and they should not only be concerned with the moral, ideological, and cultural development of teenagers but also with their mental well-being. We need to start by making the external environment better, provide in-depth psychological counseling, and support students as they work to continually increase their psychological adaptability in order to promote the healthy development of their personalities. As the epidemic situation in China has now stabilized into a normal state of prevention and control, it is imperative to provide primary and secondary school students with mental health education. In light of this context, this paper develops a browser-server network architecture-based consultation system for the mental health of students in primary and secondary schools. It eliminates the conventional booking mode and substitutes credibility as the characteristic programming parameter. The performance of the original system is improved by the reliability model the most when the parameter is set to 0.2, and the recovery rate rises by 1.32 percent. Compared to the original reliability model, which improved the system's F value performance by 0.83 percent, the accuracy rate only declines by 0.68 percent while F rises by 0.37 percent. This research is crucial for creating an information campus and raising the standard of psychological counseling.Entities:
Mesh:
Year: 2022 PMID: 36133160 PMCID: PMC9484920 DOI: 10.1155/2022/7986850
Source DB: PubMed Journal: J Environ Public Health ISSN: 1687-9805
Figure 1System architecture.
Figure 2Software functional architecture design.
Figure 3Teacher's E-R diagram.
Figure 4E-R diagram of students and systems.
Figure 5Confidence matching module class diagram.
Figure 6Clustering process.
Figure 7Influence of the number of clusters on the time of frequent template mining algorithm.
Figure 8Influence of threshold on algorithm time of frequent template mining.
Recognition result.
|
| Accuracy (%) | Recall (%) |
|
|---|---|---|---|
| Benchmark | 95.36 | 83.21 | 89.13 |
| 0.4 | 95.01 | 86.74 | 90.16 |
| 0.3 | 95.23 | 86.69 | 90.98 |
| 0.2 | 95.68 | 86.01 | 91.35 |
| 0.1 | 94.33 | 88.64 | 90.28 |
Test results of network corpus.
| Test model | Accuracy (%) | Recall (%) |
|
|---|---|---|---|
|
| 73.65 | 60.28 | 63.17 |
| Bayes | 78.91 | 62.69 | 68.83 |
| Model of this paper | 82.14 | 68.93 | 71.48 |
Figure 9MAE comparison.
Figure 10MAPE value comparison.
Figure 11Consultant clustering results.