Skin disease rates have been increasing over the past few decades. It has led to both fatal and non-fatal disabilities all around the world, especially in those areas where medical resources are not good enough. Early diagnosis of skin diseases increases the chances of cure significantly. Therefore, this work is comparing six machine learning algorithms, namely KNN, random forest, neural network, naïve bayes, logistic regression, and SVM, for the prediction of the skin diseases. The information gain, gain ratio, gini decrease, chi-square, and relieff are used to rank the features. This work comprises the introduction, literature review, and proposed methodology parts. In this research paper, a new method of analyzing skin disease has been proposed in which six different data mining techniques are used to develop an ensemble method that integrates all the six data mining techniques as a single one. The ensemble method used on the dermatology dataset gives improved result with 94% accuracy in comparison to other classifier algorithms and hence is more effective in this area.
TopIntroduction
Skin disease is one of the most important and gradually increasing disease in the world. Most of the skin diseases cases increases just my touching or getting in contact with the affected person. These types of diseases causes heavy financial, psychological and emotional burden on patients and on their families. As a result people in the world specially those who belongs to poor resources region, are unable to fight against these challenges.
Some of the common causes of skin disease includes: weakened immune system, parasites, fungus or some other microorganisms residing on the skin, many viruses, genetic factors of person, getting in get in touch with allergens, irritants, or other person’s infected skin, some bacteria those are confined in skin pores and hair follicles, diseases related to thyroid, kidneys are some important causes of dermatology diseases. Climate change may also cause skin disease but it differs from one region to another region.
Despite the fact that dermatology diseases are characterized by large range of diseases, bulk of skin diseases belong to less than ten categories as suggested by many prevalence surveys. These results are very useful in spreading information and provide preventive programs on health, which is very essential. These diseases are also categorized as infectious (that spreads from person to person) or non-infectious.
There are a variety of symptoms of skin diseases. Some of the symptoms that seem to be due to common problem, might not be every time caused due to any skin disorder. Such type of common skin problems are sores from new shoes or roughness from tight pants. Typical indications of a skin disorder include: too much flushing, raised up bumps which have red or white color, a loss of skin pigment, a rash, which might be hurting or itchy, changes in mole color or size, scaly or rough skin, bumps that are fleshy, warts, or other skin growths, ulcers, peeling skin, discolored patches of skin, open sores or lesions and cracked skin.
Many skin diseases are not avoidable, which occurs due to some genetic conditions or any other ailments. These diseases are also categorized as infectious (that spreads from person to person) or non-infectious. However, it is possible to avert the spread of skin diseases.
Some tips to prevent these types of skin conditions are:
- •
Regularly washing hands with soap and water
- •
Taking nutritious diet
- •
Avoiding sharing utensils and drinking glasses with others
- •
Getting proper vaccines to prevent some skin disorders, like chickenpox
- •
Avoiding direct contact with people who have skin infections
- •
Avoiding excessive emotional or physical stress
- •
Cleaning stuffs in public spaces before using them such as gym equipment
- •
Avoid sharing of personal things like blankets, hairbrushes, or swimsuits
- •
Drinking plenty of water
- •
Sleeping for at least seven hours each night.
TopLiterature Review
We live in an era where data is new oil. Medical data of a large number of people are collected and analyzed for medical research. Automatic diagnosis helps doctors to precisely predict the diseases. Elngar (1998) used Naïve Bayesian theorem to obtain data patterns. It helps to find the chance of happening of different dermatology diseases and the percentage of the disease has occurred.
Bapko (2011) used (Artificial Neural Networks) ANN for analysis and diagnosis of many skin infections and they attained an accuracy of 90%.