مقاله انگلیسی رایگان در مورد سیستم فازی ژنتیکی عصبی برای درجه بندی افسردگی – الزویر ۲۰۱۸

مقاله انگلیسی رایگان در مورد سیستم فازی ژنتیکی عصبی برای درجه بندی افسردگی – الزویر ۲۰۱۸

 

مشخصات مقاله
انتشار مقاله سال ۲۰۱۸
تعداد صفحات مقاله انگلیسی ۸ صفحه
هزینه دانلود مقاله انگلیسی رایگان میباشد.
منتشر شده در نشریه الزویر
نوع مقاله ISI
عنوان انگلیسی مقاله Genetic-neuro-fuzzy system for grading depression
ترجمه عنوان مقاله سیستم فازی ژنتیکی عصبی برای درجه بندی افسردگی
فرمت مقاله انگلیسی  PDF
رشته های مرتبط مهندسی کامپیوتر، روانشناسی
گرایش های مرتبط مهندسی الگوریتم ها و محاسبات، هوش مصنوعی، روانشناسی عمومی
مجله محاسبات و انفورماتیک کاربردی – Applied Computing and Informatics
دانشگاه Department of Mechanical Engineering – National Institute of Technology – India
کلمات کلیدی مدل سازی افسردگی، سیستم عصبی فازی، الگوریتم ژنتیک، تنظیم خودکار، شناسایی قوانین انحصاری
کلمات کلیدی انگلیسی Depression modeling, Neuro-fuzzy system, Genetic algorithm, Automated tuning, Identification of redundant rules
کد محصول E7512
وضعیت ترجمه مقاله  ترجمه آماده این مقاله موجود نمیباشد. میتوانید از طریق دکمه پایین سفارش دهید.
دانلود رایگان مقاله دانلود رایگان مقاله انگلیسی
سفارش ترجمه این مقاله سفارش ترجمه این مقاله

 

بخشی از متن مقاله:
۱٫ Introduction

WHO says ‘‘depression is a common illness worldwide” and it is a growing risk [1–۳]. There could be several factors contributing to depression and it is closely related to physical health [1–۳]. There exist known and effective treatment procedures, which reach less than half of the sufferers [1,4]. There are several reasons such as lack of trained professionals and lack of resources [1], silent progression [5], social stigma [6] and perception difference among doctors leading to ‘under’ or ‘over’ diagnoses [7]. So, diagnosis has to be proper to reduce the disease load [8]. SC techniques are applied for screening and diagnosing the different ailments [9,10] in the last couple of decades. Clustering techniques alone are applied for grading of depression [11]. However, Chattopadhyay et al. [12] combined fuzzy logic with clustering for capturing the symptoms psychiatric diseases of human being. They later developed a NN-based toolbox for grading of adult depression [13] and classified it into three categories namely, ‘mild’, ‘moderate’, and ‘severe’ [۱۳]. Tai and Chiu [14] used RBFN to understand the reasons for suicidal tendencies of Taiwanese soldiers. Chattopadhyay [15] developed a fuzzy-based automated model for grading depression. Later stage, Chattopadhyay et al. [16] tried to minimize the overlapping among the three different grades of depression. Regression analysis is also used for the similar purpose by Chattopadhyay and Acharya [17]. It is understood from the literature that grading of depression is a difficult task. It is mainly because of the non-availability of the data. It may of different reasons which are indicated below. (i). social stigma, (ii). majority of the patients do not approach to psychiatrists first, they approach to general physicians and when it becomes severe then comes to the psychiatrists, (iii). non existence of systems of collecting data by the doctors. For the first time reporting to the doctors, majority of the patients are prescribed medicines. As a result true picture of the patients are not available to the doctors, (iv). inferior health management system. Therefore, basic purpose of our study is to develop a prototype tool which forecasts depression better than the general physicians. In this context, SC-based techniques are found to provide formidable solutions. However, performance of SC-based techniques depends on its formulation and optimization of the same. Out of different SC-based models, neuro-fuzzy models are finding wide applications in the recent times in different field with high prediction accuracies [18,19]. Development of a neuro-fuzzy-based software prototype model has been made in this study. It has been observed that, the performance of neuro-fuzzy systems depends on its DB and RB [20]. In majority of the case, user defined the same, which in no sense becomes optimal. Some researchers tried to combine different SC-based techniques. Therefore, the main contribution of this work is the development of a neuro-fuzzybased prototype model for grading depression. Later, performance of the developed approach was made through two optimization methods, namely back-propagation algorithm and genetic algorithm. The model will be used by the general physicians for collecting more data. After proper validation with large set of data, it will be given for use to the psychiatrists/experts of mental health problems. Presently, the model is validated using data given by the clinicians. However, the model is evolutionary in nature and thus, it might handles uncertainty, impreciseness present (which are predominant) in the system in much broader way. The developed neuro-Fuzzy model and its optimization procedures using two approaches are described in Section 2 and results have been analyzed in Section 3. Final outcomes of the work are discussed and future directions are mentioned in Section 4.

برچسب ها

ثبت دیدگاه