Loading...
Please wait, while we are loading the content...
Similar Documents
An Efficient Spectral Clustering on Social Network Unstructured Data for Improved Clustering Accuracy
| Content Provider | Semantic Scholar |
|---|---|
| Author | Khanchana, R. Muthulakshmi, Paulpandi |
| Copyright Year | 2016 |
| Abstract | The paper concentrated to develop a scalable and efficient spectral clustering algorithm. It operates on big data processing in orientation of data mining approaches . Data is obtained from sensors, media sites, social media et c. A required big data is collected to find interesting patterns using data mining approaches. We collected from user links suc h as replies, posting, and retweets. We propose a probability mod el of the mentioning behavior of a social network user, to de t ct the emergence of a new topic from the anomalies measure d through the model. Aggregating anomaly scores from hundreds of user shows an emerging topics only based on the reply/me ntion relationships in social-network posts. The experime nts show that the proposed mention-anomaly-based approaches can d etect new topics at least as early as text-anomaly-based appr oaches, and in some cases much earlier when the topic is poorly id entified by the textual contents in posts in Laplacian matrix. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.ijetcse.com/wp-content/plugins/ijetcse/file/upload/docx/355AN-EFFICIENT-SPECTRAL-CLUSTERING-ON-SOCIAL-NETWORK-UNSTRUCTURED-DATA-FOR-IMPROVED-CLUSTERING-ACCURACY-pdf.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
