Normalization based K means Clustering Algorithm

K-means is an effective clustering technique used to separate similar data into groups based on initial centroids of clusters. In this paper, Normalization based K-means clustering algorithm(N-K means) is proposed. Proposed N-K means clustering algorithm applies normalization prior to clustering on the available data as well as the proposed approach calculates initial centroids based on weights. Experimental results prove the betterment of proposed N-K means clustering algorithm over existing K-means clustering algorithm in terms of complexity and overall performance.

Aspect Based Sentiment Analysis to Extract Meticulous Opinion Value

Opinion Mining and Sentiment Analysis is a process of identifying opinions in large unstructured/structured data and then analysing polarity of those opinions. Opinion mining and sentiment analysis have found vast application in analysing online ratings, analysing product based reviews, e-governance, and managing hostile content over the internet. This paper proposes an algorithm to implement aspect level sentiment analysis. The algorithm takes input from the remarks submitted by various teachers of a student. An aspect tree is formed which has various levels and weights are assigned to each branch to identify level of aspect. Aspect value is calculated by the algorithm by means of the proposed aspect tree. Dictionary based method is implemented to evaluate the polarity of the remark. The algorithm returns the aspect value clubbed with opinion value and sentiment value which helps in concluding the summarized value of remark.

Sentiment Analysis Using Collaborated Opinion Mining

Opinion mining and Sentiment analysis have emerged as a field of study since the widespread of World Wide Web and internet. Opinion refers to extraction of those lines or phrase in the raw and huge data which express an opinion. Sentiment analysis on the other hand identifies the polarity of the opinion being extracted. In this paper we propose the sentiment analysis in collaboration with opinion extraction, summarization, and tracking the records of the students. The paper modifies the existing algorithm in order to obtain the collaborated opinion about the students. The resultant opinion is represented as very high, high, moderate, low and very low. The paper is based on a case study where teachers give their remarks about the students and by applying the proposed sentiment analysis algorithm the opinion is extracted and represented.