Course Content

Abstract:

In present day Autism Spectrum Disorder (ASD) is gaining its momentum faster than ever. Detecting autism traits through screening tests is very expensive and time consuming. With the advancement of artificial intelligence and machine learning (ML), autism can be predicted at quite early stage.

Autism also called as Autism spectrum disorder (ASD) is a complex, complicated and lifelong development disability which includes problem that are characterized by repetitive behavior, non-verbal communication, lack of concentration. In recent years, ASD is increasing at a higher momentum which needs early diagnosis. Detecting Autism through various Screening tool are very time consuming and costly. In last few year, various mathematical models also called as predictive analytics are widely used for predictions. For medical science, Machine learning and pattern recognition are various multidisciplinary research areas which provide effective techniques to diagnose ASD.

 

So we proposed a system with the help of machine learning techniques and algorithms like Logistic Regression, KNN, Random Forest, Decision Tree , XGB Classifier and  Naïve Bayes to predict Autism Spectrum Disorder based on different parameters entered by the user in the front end.

Introduction:

Autism is an integral neurodevelopment disorder that affects the functioning of brain. It can occur at any age but usually occur in childhood; mostly children at the age of 2 or 3 have more chances of having ASD[1]. It occurs due to the combination of both genetic and environmental factors. Autism is not an illness or a disease rather it is a neurological condition in which child is unable to concentrate, think, learn, focus and solve the problems. They find difficulty in explaining things through facial expression or by making Gestures. It is fastest growing development disability in all over the world including India. According to Autism Centre for Excellence(ACE) 1 in 68 children is suffering for ASD. So, there is need to diagnose this disability at an early age. According to Paul Fergus, Autism, a lifelong disability is categorized in 3 ways, They are: Autistic ailment, Asperger ailment and Pervasive development ailment (PDA) also denoted as low, medium and high disorder. The child suffering from ASD faces various challenges like

1. There are various symptoms of Autism Lack of concentration 2. Repetition of same word again and again 3. Doesn’t interact with other people when they say something 4. Lack of Understanding in Making Gestures, Facial expression 5. Very sensitive in feel, touch, speech or smell. 6. Abnormal voice tone and body postures There are various causes which lead to the occurrence of ASD in child. Some of them are: Genetic- either of parents is suffering from this disorder, some other member in the family is autistic, and Complications during Pregnancy, child has not received proper and timely vaccination and so on.

 

So we proposed a system with the help of machine learning techniques and algorithms like Logistic Regression, KNN, Random Forest, Decision Tree , XGB Classifier and  Naïve Bayes to predict Autism Spectrum Disorder based on different parameters entered by the user in the front end.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Objective:

The main aim of this project to predict the Autism Spectrum Disorder using machine learning techniques and algorithms like Logistic Regression, KNN, Random Forest, Decision Tree, XGB Classifier and Naïve Bayes based on different parameters entered by the user in the front end.

Problem Statement

Autism Spectrum Disorder (ASD) is a serious developmental abnormality that seriously affects the behavior and communication of an individual. It limits the use of communicative, social and cognitive skills as well as abilities of the affected personality whereas its symptoms may vary from person to person and lifelong development disability which includes problem that are characterized by repetitive behavior, non-verbal communication, lack of concentration. In recent years, ASD is increasing at a higher momentum which needs early diagnosis. Detecting Autism through various Screening tool are very time consuming and costly.

 

Proposed System:

 

We proposed a system with the help of machine learning techniques and algorithms like Logistic Regression, KNN, Random Forest, Decision Tree, XGB Classifier and Naïve Bayes to predict Autism Spectrum Disorder based on different parameters entered by the user in the front end.