<p class="keywords">In clinical data, we have a large set of diagnostic feature and recorded details of patients for certain diseases. In a clinical environment a doctor reaches a treatment decision based on his theoretical knowledge, information attained from patients, and the clinical reports of the patient. It is very difficult to work with huge data in machine learning; hence to reduce the data, feature reduction is applied. Feature reduction has gained interest in many research areas which deals with machine learning and data mining, because it enhances the classifiers in terms of faster execution, cost-effectiveness, and accuracy. Using feature reduction we intend to find the relevant features of the data set. In this paper, we have analyzed Modified GA (MGA), PCA and combination of PCA and Modified Genetic algorithm for feature reduction. We have found that correctly classified rate of combination of PCA and Modified Genetic algorithm higher compared to other feature reduction method.</p>

A multimodal emotion recognition system is proposed using speech and facial images. For this purpose a video database is developed, containing emotions in three affective states viz. anger, sad and happiness. The audio and the snapshots of facial expressions acquired from the videos constituted the bimodal input for recognizing emotions. The spoken sentences in the database included text dependent as well as text independent sentences in Malayalam language. The audio features included short-time processing of speech to obtain: energy, zero crossing count, pitch and Mel Frequency Cepstral Coefficients. For facial expressions, the landmark features of face: eyebrows, eyes and mouth, obtained using Viola Jones Algorithm is used. The supervised learning methods K-Nearest Neighbor and Artificial Neural Network are used for emotion analysis. The system performance is evaluated for 3 cases viz. using audio based features and facial features separately and for both features taken together. Further, the effect of text dependent and text independent audio is also analyzed. The result of the analysis shows that text independent videos (utilizing both modalities) using K-Nearest Neighbor (highest accuracy 82.78%) is found to be more effective in recognizing emotions from the database considered.