Volume 18, No. 6, 2021
Classification Of Skin Disease Using LSTM With Feature Selection Method
Dr. V.Umadevi , S.Mohan
Abstract
Objective: Skin diseases are an important public health issue that affects a large number of individuals across the world. As a result, it is vital to forecast the disease sooner in order to prevent the disease from progressing to a more critical stage. In recent years, the advancement of dermatological predictive categorization has grown more predictive and accurate, thanks to fast technological advancements and the implementation of various deep learning approaches. Methods: In this work, diagnosis of skin diseases using Long-Short Term Memory is proposed which is one of the best deep learning techniques used for classification. Here, the informative Dermatology data is used to analysis and classify the skin disease with the help of the LSTM. The proposed work comprises of three steps. The procedure begins with data preprocessing, which prepares the dataset for classification. In addition, the novelty has been made in the feature selection process which is the second step of the proposed work. The recursive feature reduction approach is employed to select 10 key characteristics that play a key role in prediction. Final step is classification which is performed by the LSTM. It consists of six layers. Conclusion: When compared to other classifier methods, the presented work on Dermatology datasets gives better outcomes. From the observation, the suggested research provides a more accurate prediction result and effective way to anticipate skin diseases.
Pages: 1857-1871
Keywords: LSTM, Skin diseases, deep learning