| Literature DB >> 35615552 |
Tian Xia1.
Abstract
For the full rank of appraisement, college students act a central party in the instructive fabric of colleges and universities. The common attribute teaching should settle the reciprocal expert valuation agreeing to the specifying goals of training. Establishing a practical and energetic system for appraising the extensive rank of college students is a topical valuable for investigation. This writing confers a mandate supported on Analytic Hierarchy Process (AHP) to exactly rank the compendious degree, college students, frame estimation indicators, calculative crushing, and generate rising wherefore to distinct mayor leagues. Taking the full attribute valuation of electronic computer greater combine, a college as a represent, the import and implementation of this precept are utter details. Through the analysis of the passable state of the thorough nature appraisement, college students, alluring Taiyuan University of Science and Technology as an instance, a large property valuation dummy was established with the assistance of analytic hierarchy outgrowth. A reasonable valuation of students foresees a notional base.Entities:
Mesh:
Year: 2022 PMID: 35615552 PMCID: PMC9126681 DOI: 10.1155/2022/9606741
Source DB: PubMed Journal: Comput Intell Neurosci
Figure 1The pipeline of our method.
Comparison results of different data set before and after using optimization (data set 3).
| Set 1 | Set 2 | Set 3 | Set 4 | |
|---|---|---|---|---|
| Before | 0.5554 | 0.6657 | 0.6768 | 0.6435 |
| After | 0.6231 | 0.7004 | 0.7113 | 0.7043 |
Figure 2The pipeline of the AHP algorithm.
Comparison results of different data set before and after using optimization (data set 4).
| Set 1 | Set 2 | Set 3 | Set 4 | |
|---|---|---|---|---|
| Before | 0.5342 | 0.6342 | 0.6435 | 0.5667 |
| After | 0.6132 | 0.7032 | 0.6768 | 0.6121 |
Test accuracies of different algorithms on our adopted data set.
| Mode1 (%) | Mode2 (%) | Mode3 (%) | Mode4 (%) | |
|---|---|---|---|---|
| [ | 54.332 | 64.534 | 65.465 | 66.564 |
| [ | 64.543 | 67.668 | 70.556 | 70.557 |
| [ | 71.213 | 71.342 | 72.343 | 72.435 |
| Ours | 76.668 | 76.779 | 77.557 | 77.687 |
Comparison results of different data set before and after using optimization (data set 1).
| Set 1 | Set 2 | Set 3 | Set 4 | |
|---|---|---|---|---|
| Before | 0.4454 | 0.6231 | 0.6768 | 0.7435 |
| After | 0.6325 | 0.7435 | 0.7231 | 0.8112 |
Comparison results of different data set before and after using optimization (data set 2).
| Set 1 | Set 2 | Set 3 | Set 4 | |
|---|---|---|---|---|
| Before | 0.5768 | 0.5231 | 0.5435 | 0.6132 |
| After | 0.6214 | 0.7121 | 0.7087 | 0.7112 |
Standard errors of different algorithms on our adopted data set.
| Mode1 | Mode2 | Mode3 | Mode4 | |
|---|---|---|---|---|
| [ | 0.0657 | 0.0576 | 0.0776 | 0.0786 |
| [ | 0.0443 | 0.0546 | 0.0675 | 0.0665 |
| [ | 0.0576 | 0.0354 | 0.0657 | 0.0876 |
| Ours | 0.0043 | 0.0032 | 0.0025 | 0.0054 |
Test accuracies of different algorithms on our adopted data set.
| Mode1 (%) | Mode2 (%) | Mode3 (%) | Mode4 (%) | |
|---|---|---|---|---|
| [ | 65.454 | 65.446 | 63.435 | 65.446 |
| [ | 69.879 | 67.675 | 68.778 | 68.786 |
| [ | 65.447 | 69.786 | 70.342 | 72.331 |
| Ours | 78.675 | 84.444 | 76.557 | 76.557 |
Standard errors of different algorithms on our adopted data set.
| Mode1 | Mode2 | Mode3 | Mode4 | |
|---|---|---|---|---|
| [ | 0.0684 | 0.0786 | 0.0657 | 0.0687 |
| [ | 0.0556 | 0.0657 | 0.0879 | 0.0732 |
| [ | 0.0674 | 0.0574 | 0.067 | 0.0546 |
| Ours | 0.0032 | 0.0057 | 0.0033 | 0.0054 |