DESIGN SENSITIVE PERSONAL INFORMATION DETECTION AND CLASSIFICATION MODEL FOR AMHARIC TEXT

AMARE, GENETU MULUGETA

dc.contributor.author	AMARE, GENETU MULUGETA
dc.date.accessioned	2021-09-21T12:29:55Z
dc.date.available	2021-09-21T12:29:55Z
dc.date.issued	2021-07
dc.identifier.uri	http://ir.bdu.edu.et/handle/123456789/12621
dc.description.abstract	Sensitive information is a classified type of content that should not be disclosed to the public. And that would harm the owner of the information if it is disclosed. The collaboration between organizations, institutions, and individuals leads to a huge amount of information generation and sharing. Sharing of such information will come with disclose of sensitive information. Sensitive information can be organizational, governmental, or personal. However, whatever its type is, it needs proper handling and protection. To protect disclose of sensitive information, first it requires detecting the availability of sensitive information and its domain classification for further analysis. To the best of our knowledge, there is no work attempted for Amharic texts. Models developed for another language cannot be used for Amharic texts language because of morphology, grammar, and semantics differences. To address these gaps, we have developed a model for detecting and classifying sensitive personal information for Amharic texts. We proposed Bi-LSTM based model for sensitive information detection and classification. We have experimented with the three deep learning algorithms: LSTM, BI-LSTM, and CNN using 7.31K and 6.697K Amharic sentences for sensitivity detection and domain classification respectively on three experiments. The accuracy of LSTM, BI-LSTM, and CNN was 84%, 91%, and 88% for sensitivity classification and 88%, 93%, and 89% for domain classification respectively. The proposed Bi-LSTM model outperforms Both CNN and LSTM.	en_US
dc.language.iso	en_US	en_US
dc.subject	computer science	en_US
dc.title	DESIGN SENSITIVE PERSONAL INFORMATION DETECTION AND CLASSIFICATION MODEL FOR AMHARIC TEXT	en_US
dc.type	Thesis	en_US