Fuzzy Factor Evaluation Research

Comprehensive evaluation system refers to the number of attributes to describe the complex structure or system for the collection of relevant information, objectively evaluate its overall functioning condition (good, bad) process. Its theory and methods have become economic management, industrial engineering and decision-making indispensable and important fields.Compare comprehensive evaluation of existing, the existence of the following deficiencies : (1) the establishment of the indication system is not reasonable to study, usually in the indication system has been established for integrated evaluation methods mathematical model; (2) evaluation of existing integrated evaluation methods are being targeted to solve the classification problem, without comprehensive evaluation of various methods of research indication; (3) comprehensive evaluation of existing methods of measurement indication are qualitative and quantitative, qualitative to quantitative transformation, even if it is difficult to determine the quantitative description of the volume, thus the data obtained with information ambiguity.To solve these deficiencies, the paper presented a better, more rational, more comprehensive evaluation of the scientific method — fuzzy factors comprehensive evaluation method. Since the acquisition of data with ambiguity, the first to use attached degrees and fuzzy evaluation matrix, a multi-level transformation fuzzy more comprehensive assessment of the primary objectives of the evaluation, from the integrated value of the order was evaluated; Second, in evaluating the primary basis digging more detailed evaluation data and information, first attempt to use modern statistical algorithms and the EM algorithm for factor analysis calculations derive score factor, the analysis of the assessment in which the target has advantages and disadvantages.Comprehensive evaluation requires a reasonable indication system in study, the paper study a reasonable statistical methods to evaluate indication system. Using hypothesis testing of normal distribution and commonalty to check their common rationality, reuse factor analysis, the main components of the analysis, relevance coefficient, line structural relationship model again certify their legitimacy.Finally, Using fuzzy factor Analysis to evaluate teachers teaching quality. Not only from the teachers teaching quality assessment by the order, and a comprehensive analysis of teachers in specific aspects of the advantages and disadvantages; Meanwhile certification with the EM algorithm for factor analysis is consistent with the main components in results .So EM algorithm is feasible, but its stability is better, faster convergence. On this basis, and using EM algorithm detailed evaluation and comparison with traditional methods, and a more satisfactory results.In short, this proposed method is fuzzy factors comprehensive evaluation very good utility to improve integrated assessment methods, theory is of great significance.