K-word proximity search on encrypted data
Current Symmetric Searchable Encryption schemes do not fulfill the expectations of users that are used to web search engines. Although users are now able to search for multiple keywords, Boolean retrieval returns all results to a client regardless of how relevant the results are for the user. For searches in large data sets when also result sets are expected to be large, Boolean retrieval is not appropriate for users of modern information retrieval systems. In this paper, we present a SSE scheme that allows ranked retrieval on encrypted data, more specifically we enhanced highly-scalable Boolean retrieval with k-word proximity ranking. Additionally, we introduce an access control in our search engine, such that clients searching the data set will not learn anything about parts of the data set, for which they are not eligible.
K-word proximity search on encrypted data
IEEE 30th International Conference on Advanced Information Networking and Applications
| Authors: | Mark Gall and Gerd Brost |
| Year/month: | 2016/3 |
| Booktitle: | IEEE 30th International Conference on Advanced Information Networking and Applications |
| Pages: | 365-372 |
| Address: | Crans-Montana, Switzerland |
| Oranization: | IEEE |
| Fulltext: | click here |
Abstract |
|
| Current Symmetric Searchable Encryption schemes do not fulfill the expectations of users that are used to web search engines. Although users are now able to search for multiple keywords, Boolean retrieval returns all results to a client regardless of how relevant the results are for the user. For searches in large data sets when also result sets are expected to be large, Boolean retrieval is not appropriate for users of modern information retrieval systems. In this paper, we present a SSE scheme that allows ranked retrieval on encrypted data, more specifically we enhanced highly-scalable Boolean retrieval with k-word proximity ranking. Additionally, we introduce an access control in our search engine, such that clients searching the data set will not learn anything about parts of the data set, for which they are not eligible. | |
Bibtex:
@inproceedings { gall2016,author = { Mark Gall and Gerd Brost},
title = { K-word proximity search on encrypted data },
year = { 2016 },
month = { March },
booktitle = { IEEE 30th International Conference on Advanced Information Networking and Applications },
address = { Crans-Montana, Switzerland },
pages = { 365-372 },
organization = { IEEE },
url = { https://ieeexplore.ieee.org/document/7471228 },
}