S3BD: Secure semantic search over encrypted big data in the cloud

Abstract

Summary Cloud storage is a widely utilized service for both personal and enterprise demands. However, despite its advantages, many potential users with enormous amounts of sensitive data (big data) refrain from fully utilizing the cloud storage service due to valid concerns about data privacy. An established solution to the cloud data privacy problem is to perform encryption on the client‐end. This approach, however, restricts data processing capabilities (eg, searching over the data). Accordingly, the research problem we investigate is how to enable real‐time searching over the encrypted big data in the cloud. In particular, semantic search is of interest to clients dealing with big data. To address this problem, in this research, we develop a system (termed S3BD) for searching big data using cloud services without exposing any data to cloud providers. To keep real‐time response on big data, S3BD proactively prunes the search space to a subset of the whole dataset. For that purpose, we propose a method to cluster the encrypted data. An abstract of each cluster is maintained on the client‐end to navigate the search operation to appropriate clusters at the search time. Results of experiments, carried out on real‐world big datasets, demonstrate that the search operation can be achieved in real‐time and is significantly more efficient than other counterparts. In addition, a fully functional prototype of S3BD is made publicly available.

Survey on secure search over encrypted data on the cloud

“…28,34 For instance, in a hospital with encrypted medical records, a doctor may desire to search for records using the query ''heart disease.'' However, with a growing collection of documents and the emergence of big data, 52,53 the data users may not remember the exact keywords they want to retrieve, or they might want to search for documents that are more broadly related to a topic.…”

“…28,34 As we can see in the taxonomy of Figure 3, ontological semantic search can be achieved using synonym semantics, conceptual semantics, or a combination of these semantics. For example, if a user intends to search for robbery, she may be interested in seeing results about ''burglary'' or ''break in'' as well.…”

“…An index structure is generally implemented in form of a hash table, 33 meta data (markup), 18 or inverted index 28,34 where each unique keyword is mapped to an identifier for each document it appears in.…”

“…However, with a growing collection of documents and the emergence of big data, 52,53 the data users may not remember the exact keywords they want to retrieve, or they might want to search for documents that are more broadly related to a topic. 28,34 For instance, in a hospital with encrypted medical records, a doctor may desire to search for records using the query ''heart disease.'' While the doctor is interested in documents containing the exact query terms, she is also interested in documents with semantically related terms (eg, ''heart attack'' or ''chest pain'').…”

“…To resolve this problem, ontological semantic search was invented to find more meaningfully related data to the original query. 28,34 As we can see in the taxonomy of Figure 3, ontological semantic search can be achieved using synonym semantics, conceptual semantics, or a combination of these semantics. In this part, we explore these schemes; however, we first revisit some of the preliminary concepts that enable Ontological Searching.…”