Font Size: a A A

Mental Health Evaluation Based On Artificial Neural Network And Fuzzy Math

Posted on:2009-10-08Degree:MasterType:Thesis
Country:ChinaCandidate:C J LiFull Text:PDF
GTID:2178360245994172Subject:Applied Mathematics
Abstract/Summary:PDF Full Text Request
I believe that, the mathematization is the tendency of all science development , including the studies of the human psychology. In fact, in the field of modern psychology ,there has been a growing number of mathematical components, but also the use of a growing number of mathematical thinking and mathematical methods. However, we also noted that, at present, many of the mathematical methods do not strictly accurate empirical. How to use accurate, tight and extensive mathematical ideas and methods, how to make the psychology of knowledge systems and cognitive processes more precise, and rigorous empirical and mathematical, this is the interest of the author. This paper only capture a mental health evaluation of the small issues as the main content, hoping to get some breakthroughs.Generally speaking, if we want to establish a mathematical model in the study of psychology, first of all, wo shoule separat the psychological phenomena, such as perception, learning, decision-making, etc., from the complex psychological activities , and constitute a specific set .Then we use methods in the form of algebra, geometry, probability and justice or in the form of a computer program and formula ,to perform the object and relationships in the set. Since the establishment of the mathematical model, we will get some certain results through Logic reasoning or mathematics operation. If we give some explanation of the mathematical model, the result can be seen as as a forecast of the experience system.Using mathematical model to describe psychological phenomenon, the advantages is not only that it is more generally, accurate, and has more deductive strength and more forecast strength,more importantly, it Is advantageous for the computer simulation and it creat the condition for the artificial intelligence development.We know that this research topic - mental health assessment is essentially a pattern recognition or non-linear classification. Each individual's psychological state are multidimensional information systems, with the basic characteristics of multi-variable, multi-level, strong coupling. Between the factors there are complex non-linearity affect .On the mental health assessment, it is difficult using traditional mathematical description, but more suitable for the System analysis to consider the input and output relations. This is precisely the characteristics of artificial neural network.On the other hand, the factors impacting mental health all have fuzziness. When we determine the the impact of the informations coming from in the psychometry on the psychological health of the extent, it is difficult to have a stringent standards. When distinguishing between mental health, psychological health, as well as mental illness, the relative is also have a certain ambiguity. Therefore, we introduce the theory of fuzzy math is very feasible.Based on the above considerations, this paper will propose the sorter model to assess mental health, combining the artificial neural network and fuzzy math theory .By choosing the important factors as input vector, we use the sorter model to evaluate the psychologically healthy condition.First of all, this paper introduce the knowledge of mathematics in psychology research,By analysing the characteristics of the phenomenon study itself, we find the advantages of neural network and fuzzy mathematical in establishing the model.After the introduction of neural network and fuzzy mathematics,we immediately set up an neural network model, with conventional performance (duration), personality, character as the three input vector,which have the most impact on mental health . Using matlab neural network toolbox to train the samples,we get more precise output.In practical application, the paper further use fuzzy math methods to transform the evaluation of the three aspects -performance of the conventional, character, personality , such as "very good, good, average, poor, very poor" into a quantitative evaluation - degree of membership ,as the input data of the model to diagnose a person's mental state. Compared to the traditional method based on complete experience,this new method improves the accuracy of the mathematical model. Also, it is much more simple and More scientific...
Keywords/Search Tags:artificial neural networks, fuzzy math, mathematical model, Perceptron, psychological assessment
PDF Full Text Request
Related items