AbstractsComputer Science

In this thesis a new intelligent temporal mining system has been newlineproposed and implemented to perform effective social network analysis For newlinethis purpose two new clustering algorithms namely neuro genetic weighted newlinefuzzy c means clustering algorithm and hybrid clustering based classification newlinealgorithm are proposed in this thesis In addition, a new seven factor analysis newlinemodel has been designed and implemented These algorithms are capable of newlineselecting the optimal number of clusters from the social network data sets newlineThese clustering algorithms are used to perform classification of social newlinenetwork users Moreover these algorithms reduce the classification time newlineThese two new clustering algorithms have been proposed and newlineimplemented for selecting the optimal number of clusters necessary for newlineanalyzing and improving the classification accuracy Based on these a new newlineuser behavior analysis model is developed using the proposed Neuro Genetic newlineWeighted Fuzzy C Means Clustering Algorithm NGWFCMA by effective newlineClustering The proposed clustering algorithm helps to improve the newlineclassification accuracy by providing initial groups Moreover a seven factor newlineanalysis is introduced in this work to help the clustering algorithm In newlineaddition Adaptive Neuro Fuzzy Inference System ANFIS algorithm is also newlineused for classifying the data The experimental results obtained from this newlinework show that the clustering algorithm helps to provide better classification newlineaccuracy when tested with Facebook dataset newlineSecond a Hybrid Clustering Based Classification Algorithm newline HCBCA has been proposed in this work to improve the classification newlineperformance by selecting the clusters newline newline%%%reference p127-141.