Keywords : Learning representives
European Journal of Molecular & Clinical Medicine,
2020, Volume 7, Issue 9, Pages 2832-2847
Text clustering is an important method for effectively organising, summarising, and navigating text information. The purpose of the clustering is to distinguish and classify the similarity among the text instance as label. However, in the absence of labels, the text data to be clustered cannot be used to train the text representation model based on deep learning as it contains high dimensional data with complex latent distributions. To address this problem, a new unified deep learning framework for text clustering based on deep representation learning is proposed using the deep adaptive fuzzy clustering in this paper to provide soft partition of data. Initially reconstruction of original data into feature space carried out using the word embedding process of deep learning. Word embedding process is a learnt representation of the text or sentence towards clustering into vector containing words, characters and N-grams of words. Further clustering of feature vector is carried out with max pooling layer to determine the inter-cluster seperability and intra-cluster compactness. Moreover learning of the feature space is processed with gradient descent. Moreover tuning of feature vector is fine tuned on basis of Discriminant information using hyper parameter optimization with fewer epochs. Finally representation learning and soft clustering has been achieved using deep adaptive fuzzy clustering and quantum annealing based optimization has been employed .The results demonstrate that the clustering approach more stable and accurate than the traditional FCM clustering algorithm on employing k fold validation for evaluation. The Experimental results demonstrates the proposed technique outperforms of state of arts approaches in terms of set based measures like Precision, Recall and F measure and rank based measures like Mean Average Precision and Cumulative Gain.