PTSSP: privacy-preserving top-k spatial keyword similarity query with priority matching

As a popular and practical query type in location-based services, the spatial keyword query has been extensively studied in both academia and industry. Meanwhile, with the growing demand for data privacy, many privacy-preserving spatial keyword query schemes have been proposed to deal with queries over encrypted data. However, none of the existing schemes preserve access pattern privacy, and the recent research illustrates that leaking such privacy may incur inference attacks and thus disclose sensitive information. In addition, most existing schemes only consider the boolean keyword search, which is not quite practical and flexible in real-world applications. To address the above issues, in this paper, we propose two privacy-preserving spatial keyword similarity query schemes that can preserve full and partial access pattern privacy, respectively. First, we present a basic privacy-preserving spatial keyword similarity query scheme (PPSKS) by integrating a secure set membership test (SSMT) technique with secure circuits. After that, to improve performance, we propose a tree-based scheme (PPSKS+) by employing a new index called FR-tree together with a predicate encryption technique that can encrypt FR-tree. Formal security analysis shows that: i) our proposed schemes can protect outsourced data, query requests, and query results; ii) our PPSKS scheme can hide full access patterns, while the PPSKS+ scheme preserves <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$m$</tex-math></inline-formula> -access pattern privacy. Extensive experiments are also conducted, and the results indicate that our tree-based PPSKS+ scheme is much more efficient, almost two orders of magnitude better than our linear search PPSKS scheme in performing queries.

Traditional spatial querying assumes that a user specifies exactly the constraints of valid results, and that the result set contains only those items that fulfill exactly the query constraints. The nature of spatial data, however, makes it difficult for a user to always guess correctly the values stored, while exhaustive enumerations of acceptable alternatives to the ideal target would become a tedious process. Likewise, values that deviate somewhat from the query constraints should be part of a ranked result set as well. This paper develops methods for the retrieval of similar spatial information, in particular when several similarity constraints with logical combinations must be compared and integrated. The first set of methods is concerned with spatial similarity reasoning over null values by denoting the semantics of different types of null values with explicit identifiers that imply different degrees of uncertainty. The second set of methods contributes to a consistent and comprehensive methodology for spatial similarity retrieval in response to complex queries with combinations of logical operators. We provide an exhaustive list of spatial query scenarios with conjunctions, disjunctions, and negation and present justified solutions for each case. Since these computational methods are founded on fundamental psychological knowledge about similarity reasoning, they are particularly well suited for generating similarity results that match with users’ intuitions.

Personalizing the Top-k Spatial Keyword Preference Query with textual classifiers

The increasing prevalence of cloud computing drives the exploration of various secure query schemes over encrypted data, among which secure spatial keyword query has drawn a great deal of attention due to its broad application in location-based services. However, most existing schemes are either limited to the boolean keyword test or incapable of protecting access pattern privacy. Although the state-of-the-art secure spatial keyword query scheme can support keyword similarity while preserving access pattern privacy, it is unable to cope with the arbitrary spatial range, which is more general, and has limitations in efficiency and security. In this paper, we propose a new secure spatial keyword similarity query scheme that can support arbitrary spatial ranges and enhance the efficiency and security of the state-of-the-art scheme at the same time. Specifically, we first present a new homomorphic encryption technique by improving the popular symmetric homomorphic encryption (SHE). After that, we propose a novel approach to make supporting arbitrary spatial ranges over encrypted data possible, in which a spatial encoding technique is designed to improve performance. Finally, by designing a pack-based solution to protect access pattern privacy, our proposed scheme can hide the number of query results while optimizing performance. We formally prove the security of our proposed scheme and conduct experiments to evaluate its performance. The results indicate that our proposed scheme outperforms the state-of-the-art scheme in both the computational costs and communication overhead.

With the proliferation of geo-positioning and geo-tagging, spatial keyword query (SKQ) processing is gaining great concern recently. Typically, a top-k spatial keyword query returns the k best spatio-textual objects ranked according to their proximity to the query location and relevance to the query keywords. While there has been much work devoted to top-k spatial keyword processing, most of them are either focused on single query or only suitable for Euclidean space. In this paper, we take the first step to study the problem of multiple query points (or group) top-k spatial keyword query processing in road networks. We first propose a basic group query-processing algorithm by using three distinct index structures. Next, we propose another more efficient algorithm based on the concept of Minimum Bounding Rectangle (MBR), which can significantly reduce the objects to be examined and thus achieve higher performance. Finally, extensive experiments using real data sets are conducted to validate the effectiveness of the proposed algorithms.

Similarity is the assessment of deviation from equivalence. Spatial scene similarity assessment is a hot topic in Geographic Information Science and has wide prospect in spatial analysis, spatial query, and map generalization, etc. Spatial scene similarity is usually employed to measure similarity between query spatial scene and various candidate scenes. The spatial relation between geographic objects are consisted of topological, directional, and distance. As they capture the essence of a scene's structure, conventional spatial scene similarity computation method builds synthetic evaluation models by adjusting coefficient among different parameters. In this paper, deep neural networks were introduced to assess the spatial scene similarity. More than 50 million map search case was used to train the model. The case data was from Tusou system, which was developed to search map via spatial scene by users' sketch. Experimental implement show that the new model understands users' intension better than current model.

Conventional top-k spatial keyword queries require users to explicitly specify their preferences between spatial proximity and keyword relevance. In this work we investigate how to eliminate this requirement by enhancing the conventional queries with interaction, resulting in Interactive Top-k Spatial Keyword (ITkSK) query. Having confirmed the feasibility by theoretical analysis, we propose a three-phase solution focusing on both effectiveness and efficiency. The first phase substantially narrows down the search space for subsequent phases by efficiently retrieving a set of geo-textual k-skyband objects as the initial candidates. In the second phase three practical strategies for selecting a subset of candidates are developed with the aim of maximizing the expected benefit for learning user preferences at each round of interaction. Finally we discuss how to determine the termination condition automatically and estimate the preference based on the user's feedback. Empirical study based on real PoI datasets verifies our theoretical observation that the quality of top-k results in spatial keyword queries can be greatly improved through only a few rounds of interactions. © 2015 IEEE.

Spatial Scene Similarity Assessment (SSSA) is an essential problem in spatial analysis, spatial query, and map generalization, etc. In SSSA, spatial scene similarity needs to be compared between query spatial scene and each candidate spatial scene. The computational complexity of spatial scene comparison often cannot be resolved by sequential computing model. In this paper, we analyze the computational complexity of SSSA and develop a parallel processing method and associated algorithms for SSSA based on Hadoop. The COOT (Cell Object Overlay Times) is proposed as a data locality strategy. The experiment results demonstrate that MapReduce on Hadoop significantly improve SSSA in computing performance and data processing capability.

Computational similarity assessments over spatial objects are typically decomposed into similarity comparisons of geometric and non-geometric attribute values. Psychological findings have suggested that different types of aggregation functions—for the conversions from the attributes’ similarity values to the objects’ similarity values—should be used depending on whether the attributes are separable (which reflects perceptual independence) or whether they are integral (which reflects such dependencies among the attributes as typically captured in geometric similarity measures). Current computational spatial similarity methods have ignored the potential impact of such differences, however, treating all attributes and their values homogeneously. Through a comprehensive simulation of spatial similarity queries the impact of psychologically compliant (which recognize groups of integral attributes) vs. deviant (which fail to detect such groups) methods have been studied, comparing the top-10 items of the compliant and deviant ranked lists. We found that only for objects with very small numbers of attributes—no more than two or three attributes for the objects—the explicit recognition of integral attributes is negligible, but the differences between compliant and deviant methods become progressively worse as the percentage of integral attributes increases and the number of groups in which these integral attributes are distributed decreases.

Many real-time applications use spatial keyword queries, which provide location information and text descriptions of Points Of Interest (POI). Since a promising research subject can pose such a procedure, performing spatial keyword queries over encrypted data is difficult. Several schemes for addressing secure spatial and keyword queries have been proposed, but previous surveys have only summarised and tested secure query algorithms. They still lack a privacy-preserving overall review of the spatial keyword query. This paper outlines a secure spatial keyword query’s three main components: secure spatial query, secure textual query, and secure spatial-textual query. Some evaluation criteria have been carefully selected to aid in the evaluation of existing spatial keywords. Following that, we expand on and evaluate recent related research’s benefits and disadvantages using the suggested criteria. Finally, we see some unanswered problems that researchers can use to perform more analyses and studies.

Spatial queries are becoming integral part of many mobile applications. With revolutions in GPS and GIS, the location based spatial queries are exploring spatial information to mobile users to a large extent. Finding nearest neighbor, shortest path to the destination, retrieving skyline objects and spatial skyline keyword query are some of commonly used queries on road networks. In this paper we introduce a new query type called Boolean Skyline Boundary (BSB) query on road networks. The algorithm retrieves the BSB objects based on both network distance and travel time on road networks. The algorithm searches the objects based on user specified spatial boundary and the Boolean skyline expression. The algorithm also specifies the shortest path to reach the BSB objects. The experimental evaluation reflects the performance of BSB algorithm.

Time-aware spatial keyword queries in traffic networks ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TSKQT</i> ) aim to retrieve top- <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">k</i> result objects based on the ranking score considering spatial proximity, textual relevance, and temporal similarity simultaneously. However, due to inappropriate query parameter settings, the query result set may contain one or more objects not expected by the user. To improve the usability of query results, we propose and study the why question on time-aware spatial keyword queries in the traffic network ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">WhyTSKQT</i> ) for the first time. Specifically, a hybrid index structure, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">TTG-tree</i> , is proposed to effectively organize the traffic network information and the positional, textual, and temporal information of objects. Moreover, several pruning strategies are presented to filter out massive objects irrelevant to the query. By analyzing the keywords contained by original result objects and studying the relationship between the time intervals of why objects and the query time interval, we can reasonably construct the candidate query keyword set and query time set to form the candidate refined queries, so as to minimize the refined queries to be evaluated. In addition, several optimization techniques are proposed to further speed up the acquisition of the lowest-cost refined query. Finally, extensive experiments are carried out on two traffic networks to verify the efficiency of the proposed method.

Recently, Collective Spatial Keyword Querying (CoSKQ), which returns a group of objects that cover a set of given keywords collectively and have the smallest cost, has received extensive attention in spatial database community. However, no research so far focuses on a situation when the result of CoSKQ is taken as the input of a query. But this kind of query has many applications in location based services. In this paper, we introduce a new problem Reverse Collective Spatial Keyword Querying (RCoSKQ) that returns a region, in which the query objects are qualified objects with the highest spatial and textual similarity. We propose an efficient method which uses IR-tree to retrieve objects with text descriptions. To accelerate the query process, a pruning method that effectively reduces computing is proposed. The experiments over real and synthesis data sets demonstrate the efficiency of our approaches.

Spatial‐query‐by‐sketch is an intuitive tool to explore human spatial knowledge about geographic environments and to support communication with scene database queries. However, traditional sketch‐based spatial search methods perform inadequately due to their inability to find hidden multiscale map features from mental sketches. In this research, we propose a deep convolutional neural network, namely the Deep Spatial Scene Network (DeepSSN), to better assess the spatial scene similarity. In DeepSSN, a triplet loss function is designed as a comprehensive distance metric to support the similarity assessment. A positive and negative example mining strategy is designed to ensure a consistently increasing distinction of triplets during the training process. Moreover, we develop a prototype spatial scene search system using the proposed DeepSSN, in which the users input spatial queries via sketch maps and the system can automatically augment the sketch training data. The proposed model is validated using multisource conflated map data including 131,300 labeled scene samples after data augmentation. The empirical results demonstrate that the DeepSSN outperforms baseline methods including k‐nearest neighbors, the multilayer perceptron, AlexNet, DenseNet, and ResNet using mean reciprocal rank and precision metrics. This research advances geographic information retrieval studies by introducing a novel deep learning method tailored to spatial scene queries.

Spatial approximate keyword queries consist of a spatial condition and a set of keywords as the fuzzy textual conditions, and they return objects labeled with a set of keywords similar to queried keywords while satisfying the spatial condition. Such queries enable users to find objects of interest in a spatial database, and make mismatches between user query keywords and object keywords tolerant. With the rapid growth of data, spatial databases storing objects from diverse geographical regions can be no longer held in main memories. Thus, it is essential to answer spatial approximate keyword queries over disk resident datasets. Existing works present methods either returns incomplete answers or indexes in main memory, and effective solutions in disks are in demand. This paper presents a novel disk resident index RMB-tree to support spatial approximate keyword queries. We study the principle of augmenting R-tree with capacity of approximate keyword searching based on existing solutions, and store multiple bitmaps in R-tree nodes to build an RMB-tree. RMB-tree supports spatial conditions such as range constraint, combined with keyword similarity metrics such as edit distance, dice etc. Experimental results against R-tree on two real world datasets demonstrate the efficiency of our solution.