|
| Method category | Calculation | Advantages and disadvantages | Example |
|
| Subjective empowerment method | The weights are obtained by industry experts through analysis and processing based on experience. | Relying on subjective experience, the relative size of the weights is less likely to be wrong, but it cannot obtain high numerical accuracy. | Expert scoring method, analytic hierarchy process, etc. |
|
| Objective empowerment | According to the relationship and characteristics contained in a large amount of data in the case database, the weight is calculated from the actual value. | The weights with finer precision can be obtained without relying on human subjective experience, but the quality requirements of the original data are high, and the relative size of the weights cannot be guaranteed to be correct. | Rough set theory method, weight method, dispersion method, etc. |
|
| Combination of subjective and objective | Combining the results of the subjective and objective weighting methods in a certain proportion or modifying the results of a single subjective or objective method by another. | While ensuring the relative size of the weights is correct, the influence of subjective factors is weakened to a certain extent. A more reasonable fusion method needs to be further explored. | — |
|
| Artificial intelligence method | According to the characteristics of the characteristic attributes, use various intelligent algorithms that imitate the way of human thinking or the algorithms obtained by modeling and summarizing certain types of practical problems to calculate the attribute weights. | The weight calculation model is further complicated, and the diversified artificial intelligence technology also provides a more suitable calculation model for the weight calculation in various fields, making the calculation results more objective and accurate. | Genetic algorithm, neural network, ant colony algorithm, particle swarm algorithm, etc. |
|
