Refine your search
Collections
Co-Authors
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Sakthivel, C. R.
- An Efficient Naive Bayes Classification for Sentiment Analysis on Twitter
Abstract Views :202 |
PDF Views:2
Authors
Affiliations
1 Department of Computer Science, Sri Ramakrishna Mission Vidyalaya, College of Arts and Science, Coimbatore-20, IN
2 Department of Computer Science, Sri Ramakrishna Mission Vidyalaya College of Arts and Science, Coimbatore-20, IN
1 Department of Computer Science, Sri Ramakrishna Mission Vidyalaya, College of Arts and Science, Coimbatore-20, IN
2 Department of Computer Science, Sri Ramakrishna Mission Vidyalaya College of Arts and Science, Coimbatore-20, IN
Source
Data Mining and Knowledge Engineering, Vol 7, No 5 (2015), Pagination: 180-185Abstract
Twitter is one of the large amounts of tweets contained social media site. That's being as a good platform for tracking and analyzing public sentiment. Nowadays the one of the important data mining research is sentiment analysis mining or opinion mining analyzing. That's why the sentiment analysis is attracting the researchers to find the critical information for decision making purpose in both of academic sides and also industry sides. This proposed system finds the sentiment variations. It implements existing FB-LDA and RCB-LDA algorithm with new Naive Bayes algorithm using C# .Net for effectively handle the class imbalance Problem of positive and negative changes made by improper sentiment label assignment also improves the accuracy significantly than the existing system.Keywords
Sentiment Analysis, FB-LDA, RCB-LDA, Opinion Mining, Tweets, Classification, Naive Bayes Etc.
- A New Top-K Nearest Neighbor and Range Combined Query Processing for Spatial Data
Abstract Views :186 |
PDF Views:2
The wide range of location-based applications that rely on spatial preference queries. Range query are used to find the accurate results from the spatial location. The essential idea is to precompute some accurate information in top K queries. To propose a novel technique to speed up the performance of top-k spatial preference queries with nearest neighbor query and range query. In proposed system, determining a range query to evaluate a top-k query by exploiting the nearest neighbor algorithm. Top k nearest range query algorithm is used to assign the range values for each nearest neighbor queries at given query. After assign the range values to the query, calculating all nearest neighbor using the nearest neighbor algorithm. Then calculating the nearest neighbor query with range values, the top k queries are selected for query result. The result of proposed new Top k nearest neighbor and range combined query processing is compared with existing nearest neighbor query techniques. Finally the performance of top k nearest range query algorithm process produce more efficient result and retrieval efficiency of the resulting scheme is high.
Authors
Affiliations
1 Department of Computer Science, Sri Ramakrishna Mission Vidyalaya College of Arts and Science, Coimbatore, Tamilnadu, IN
1 Department of Computer Science, Sri Ramakrishna Mission Vidyalaya College of Arts and Science, Coimbatore, Tamilnadu, IN
Source
Data Mining and Knowledge Engineering, Vol 6, No 6 (2014), Pagination: 252-254Abstract
Nearest Neighbor queries and All Nearest Neighbor (ANN) operation is a commonly used primitive for analyzing large multi-dimensional datasets, ANN is very expensive. The traditional index-based methods use a pruning metric called MAXMAXDIST and a new ANN pruning metric called NXNDIST, it reduces the computation time. The existing system stores the bucket quad tree index structure, called MBRQT. It is using extensive experimental evaluation to show that MBRQT index can significantly speed up the computation and efficiently answer the more general All-k-Nearest-Neighbor (AkNN) queries. Top-k spatial preference query retrieves the k best data objects in road network with highest scores. The score of an object is defined by the quality of features in its spatial neighborhood.The wide range of location-based applications that rely on spatial preference queries. Range query are used to find the accurate results from the spatial location. The essential idea is to precompute some accurate information in top K queries. To propose a novel technique to speed up the performance of top-k spatial preference queries with nearest neighbor query and range query. In proposed system, determining a range query to evaluate a top-k query by exploiting the nearest neighbor algorithm. Top k nearest range query algorithm is used to assign the range values for each nearest neighbor queries at given query. After assign the range values to the query, calculating all nearest neighbor using the nearest neighbor algorithm. Then calculating the nearest neighbor query with range values, the top k queries are selected for query result. The result of proposed new Top k nearest neighbor and range combined query processing is compared with existing nearest neighbor query techniques. Finally the performance of top k nearest range query algorithm process produce more efficient result and retrieval efficiency of the resulting scheme is high.