Abstract:
NER and human community detection use short-text author links. However, challenges await. First, the input short-text contents are noisy, ambiguous, and non-grammatical. Second, traditional text mining methods cannot extract concepts from words and phrases.
Third, the text is temporally skewed, which can affect semantic understanding in multiple ways. Finally, knowledge-bases can bias results to the external database and change the meaning of the input short text corpus. A neural network-based temporal-textual framework generates subgraphs with highly correlated authors from short-text contents to overcome these challenges.
Our method uses a portmanteau of contents and concepts to calculate the relevance score (edge weight) between authors and a stack-wise graph cutting algorithm to extract their communities.
Our multi-aspect vector space model outperforms knowledge-centered competitors in linking short-text authors, according to experiments. Given the author linking task, the more comprehensive the dataset, the more significant the extracted concepts.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
