Reverse K-nearest Neighbor Research Articles

CSD-RkNN: reverse k nearest neighbors queries with conic section discriminances

The reverse k nearest neighbors (RkNN) query is a prominent yet time-consuming spatial query used in facility siting, influential domain analysis, potential customer analysis, etc. Its aim is to identify all points that consider the query point as one of their k closest points. However, when k is relatively large (e.g. k = 1000), existing RkNN techniques often struggle to provide acceptable response times (within a few seconds). To address this issue, we propose a verification approach called conic section discriminance (CSD). This method serves to determine whether points belong to the RkNN set. With CSD, only a small fraction of candidates require costly k nearest neighbors (kNN) queries for verification, while the rest can be rapidly verified with O(1) complexity. Furthermore, we propose a Voronoi-based candidate generation approach to curtail the candidate set size. By leveraging the VoR-tree structure, we integrate these two approaches to form a novel RkNN algorithm named CSD-RkNN. A comprehensive set of experiments is conducted to compare CSD-RkNN with Slice as the state-of-the-art RkNN algorithm, and VR-RkNN as the original RkNN algorithm on VoR-tree. The results indicate that CSD-RkNN consistently outperforms the other two algorithms, especially when k is relatively large.

Efficient and Privacy-Preserving Spatial-Feature-Based Reverse kNN Query

Reverse k nearest neighbor (RkNN) query has been widely applied in the targeted push of information. Many schemes for the RkNN query on encrypted data have been proposed for coordinating the emerging trend of outsourcing data to the cloud. However, none of them supports the spatial data with many features, a prevalent data type in location-based services, e.g., each user in online dating apps usually has a spatial location and many personality trait features. Meanwhile, incorporating features with the spatial data endows the spatial-feature-based RkNN query to provide more precise services than the spatial-based RkNN query. Therefore, as a steppingstone, we propose an efficient and privacy-preserving spatial-feature-based RkNN scheme in this work for the first time. Specifically, we first design a modified intersection and union R tree (MIUR-tree) to index the spatial and feature data. Then, we introduce an MIUR-tree based RkNN query algorithm in the filter and refinement framework to efficiently process RkNN queries. After that, based on a symmetric homomorphic encryption (SHE) scheme, we design a private filter protocol and a private refinement protocol, and leverage them to propose our RkNN query scheme. Rigorous security analysis demonstrates that our scheme is privacy-preserving, and extensive experiments indicate that our scheme is computationally efficient.

Bidirectional k-nearest neighbor spatial crowdsourcing allocation protocol based on edge computing.

Spatial crowdsourcing refers to the allocation of crowdsourcing workers to each task based on location information. K-nearest neighbor technology has been widely applied in crowdsourcing applications for crowdsourcing allocation. However, there are still several issues need to be stressed. Most of the existing spatial crowdsourcing allocation schemes operate on a centralized framework, resulting in low efficiency of crowdsourcing allocation. In addition, these spatial crowdsourcing allocation schemes are one-way allocation, that is, the suitable matching objects for each task can be queried from the set of crowdsourcing workers, but cannot query in reverse. In this article, a bidirectional k-nearest neighbor spatial crowdsourcing allocation protocol based on edge computing (BKNN-CAP) is proposed. Firstly, a spatial crowdsourcing task allocation framework based on edge computing (SCTAFEC) is established, which can offload all tasks to edge nodes in edge computing layer to realize parallel processing of spatio-temporal queries. Secondly, the positive k-nearest neighbor spatio-temporal query algorithm (PKNN) and reverse k-nearest neighbor spatio-temporal query algorithm (RKNN) are proposed to make the task publishers and crowdsourcing workers conduct two-way query. In addition, a road network distance calculation method is proposed to improve the accuracy of Euclidean distance in spatial query scenarios. Experimental results show that the proposed protocol has less time cost and higher matching success rate compared with other ones.

SetRkNN: Efficient and Privacy-Preserving Set Reverse kNN Query in Cloud

The advance of cloud computing has driven a new paradigm of outsourcing large-scale data and data-driven services to public clouds. Due to the increased awareness of privacy protection, many studies have focused on addressing security and privacy issues in outsourced query services. Although many privacy-preserving schemes have been proposed for various query types, the set reverse k nearest neighbors (RkNN) query is still an unexplored area. Even if some existing schemes can be adapted to achieve privacy-preserving set RkNN queries, they will suffer from linear search efficiency. As a steppingstone, in this paper, we propose an efficient and privacy-preserving set RkNN query scheme over encrypted data with sublinear query efficiency. Specifically, we first design an inverted prefix index to organize the set dataset and propose an algorithm to traverse the index with sublinear search efficiency. Then, we propose two oblivious data comparison protocols based on a symmetric homomorphic encryption (SHE) scheme and design the private filter/refinement protocols to preserve the privacy of index searching. After that, we propose an access pattern privacy-preserving set RkNN query scheme by using private filter/refinement protocols. Rigorous security analysis demonstrates that our scheme can protect data privacy and access pattern privacy. Experimental results indicate that our scheme is more efficient than the available naive solution in terms of computational costs and communication overheads.

Robust outlier detection based on the changing rate of directed density ratio

The task of outlier detection aims at mining abnormal objects that deviate from normal distribution. Traditional unsupervised outlier detection methods can detect most global outliers, but only perform well under relatively single data distribution. Although the methods based on k-nearest neighbors can fit more complex data distribution, they also have the problem of hardly detecting local outliers or the performance easily influenced by data manifolds. At the same time, the outlier detection performance of most methods based on k-nearest neighbors is greatly affected by parameter k. We proposed a robust outlier detection method based on the changing rate of directed density ratio. The local density of samples is calculated by combining kernel density estimation and extended neighbor set which contains k-nearest neighbors and reverse k-nearest neighbors. Then we define the directed density ratio of a sample based on the density ratio and the vector between the sample and its neighbors. The local information can be better estimated by directed density ratio under different local densities and data manifolds. Then, by increasing the size of neighbors, the change of directed density ratio of a sample was calculated and finally summed up as the outlier score. Experiments are carried out on 12 synthetic datasets that simulate different data distributions and 22 public datasets. The experimental results show that compared with several state-of-the-art methods, the proposed method can achieve better outlier detection performance under different data distributions. In addition, the proposed method shows better robustness when parameter k changes in experimental results.

SMOTE-RkNN: A hybrid re-sampling method based on SMOTE and reverse k-nearest neighbors

In recent years, class imbalance learning (CIL) has become an important branch of machine learning. The Synthetic Minority Oversampling TEchnique (SMOTE) is considered to be a benchmark algorithm among CIL techniques. Although the SMOTE algorithm performs well on the vast majority of class-imbalance tasks, it also has the inherent drawback of noise propagation. Many SMOTE-variants have been proposed to address this problem. Generally, the improved solutions conduct a hybrid sampling procedure, i.e., carrying out an undersampling process after SMOTE to remove noises. However, owing to the complexity of data distribution, it is sometimes difficult to accurately identify real instances of noise, resulting in low modeling quality. In this paper, we propose a more robust and universal SMOTE hybrid variant algorithm named SMOTE-reverse k-nearest neighbors (SMOTE-RkNN). The proposed algorithm identifies noise based on probability density but not local neighborhood information. Specifically, the probability density information of each instance is provided by RkNN, a well-known KNN variant. Noisy instances are found and deleted according to their relevant probability density. In experiments on 46 class-imbalanced data sets, SMOTE-RkNN showed promising results in comparison with several popular SMOTE hybrid variant algorithms.

Collective Data Anomaly Detection Based on Reverse k-Nearest Neighbor Filtering

Collective Data Anomaly Detection Based on Reverse k-Nearest Neighbor Filtering

An evaluation and query algorithm for the influence of spatial location based on RkNN

This paper is devoted to the investigation of the evaluation and query algorithm problem for the influence of spatial location based on RkNN (reverse k nearest neighbor). On the one hand, an object can make contribution to multiple locations. However, for the existing measures for evaluating the influence of spatial location, an object only makes contribution to one location, and its influence is usually measured by the number of spatial objects in the region. In this case, a new measure for evaluating the influence of spatial location based on the RkNN is proposed. Since the weight of the contribution is determined by the distance between the object and the location, the influence weight definition is given, which meets the actual applications. On the other hand, a query algorithm for the influence of spatial location is introduced based on the proposed measure. Firstly, an algorithm named INCH (INtersection’s Convex Hull) is applied to get candidate regions, where all objects are candidates. Then, kNN and Range-k are used to refine results. Then, according to the proposed measure, the weights of objects in RkNN results are computed, and the influence of the location is accumulated. The experimental results on the real data show that the optimized algorithms outperform the basic algorithm on efficiency. In addition, in order to provide the best customer service in the location problem and make the best use of all infrastructures, a location algorithm with the query is presented based on RkNN. The influence of each facility is calculated in the location program and the equilibrium coefficient is used to evaluate the reasonability of the location in the paper. The smaller the equilibrium coefficient is, the more reasonability the program is. The actual application shows that the location based on influence makes the location algorithm more reasonable and available.

Reverse Approximate Nearest Neighbor Queries on Road Network

Reverse k Nearest Neighbor (RkNN) queries retrieve all objects that consider the query as one of their k most influential objects. Given a set of user U, a set of facilities F and a value k, a facility f is said to be influential to a user u if f is one of the k closest facilities to u. As a complement of RkNN query, Reverse Approximate Nearest Neighbor (RANN) query considers relaxed definition of influence, where a user u is influenced by not only its closest facility, but also by every other facility that is almost as close to u as its closest facility is. In this paper, we study RANN query on road network. Existing RANN techniques and algorithms only work for queries on Euclidean space and are not directly applicable for RANN queries on road network. We propose pruning techniques that utilize Network Voronoi Diagram (NVD) to efficiently solve RANN query on road network. We conduct extensive experimental study on different real data sets and demonstrate that our algorithm is significantly better than the competitor.

Adaptive density peak clustering based on dimensional-free and reverse k-nearest neighbors

Cluster analysis plays a crucial component in consumer behavior segment. The density peak clustering algorithm (DPC) is a novel density-based clustering method. However, it performs poorly in high-dimension datasets and the local density for boundary points. In addition, its fault tolerance is affected by one-step allocation strategy. To overcome these disadvantages, an adaptive density peak clustering algorithm based on dimensional-free and reverse k-nearest neighbors (ERK-DPC) is proposed in this paper. First, we compute Euler cosine distance to obtain the similarity of sample points in high-dimension datasets. Then, the adaptive local density formula is used to measure the local density of each point. Finally, the reverse k-nearest neighbor idea is added on two-step allocation strategy, which assigns the remaining points accurately and effectively. The proposed clustering algorithm is experiments on several benchmark datasets and real-world datasets. By comparing the benchmarks, the results demonstrate that the ERK-DPC algorithm superior to some state-of- the-art methods.

Can Reverse Nearest Neighbors Perceive Unknowns?

A novel open set classifier is presented in this work, where the neighborhood of a test instance is determined using the principles of Reverse k-nearest neighbors (RkNN). The RkNN count of an instance can have any non-negative value less or equal to the size of the training set. While dealing with an open dataset, consisting of known and unknown classes, the zero count can provide a possible solution for detecting the unknown class. Positive RkNN count along with the nearest RkNN distance information are used to determine the known class classifications. Experiments are carried out on ten real world datasets, with various openness values on five state-of-the-art open set learners and the proposed scheme. Their performance is measured on three evaluating metrics namely accuracy, average F1 over known and unknown classes, and Known class F1. Empirical results indicate comparable to superior performance delivered by the proposed method over the state-of-the-art approaches on all but one dataset.

Online Data Center Traffic Classification Based on Inter-Flow Correlations

Today, increasing attention is being paid to Data Center (DC) traffic classification since these infrastructures have become the heart of a variety of time-sensitive and data-intensive service platforms. Classification provides the required tools for better understanding traffic patterns in order to ensure high Quality of Service (QoS) performances and solve scalability problems. Unfortunately, existing classification algorithms cannot deal efficiently with two critical challenges in DC traffic: inter-class imbalance and critical time constraints. In this paper, we propose a novel correlation-based algorithm following a cost-sensitive approach combined with a Bagged Random Forest (BRF) ensemble algorithm, to address the inter-class imbalance problem while meeting time requirements in a data center environment. In this strategy, a new method based on Reverse k-Nearest Neighbors (RkNN) is proposed to capture the rebalancing weights expressing inter-flow correlations, in order to perform an online classification approach. We demonstrate the efficiency of the algorithm by comparing its performance to several existing methods from data level, algorithm level, and cost-sensitive strategies on four real-world datasets. The results reveal that the proposed algorithm outperforms most approaches in the different datasets in terms of precision, recall, F1 measure, AUC and Kappa, as opposed to other algorithms that result in either high precision with low recall and low precision and high recall causing congestion or resource over provisioning.

A proposition of querying scheme with network Voronoi diagram in bichromatic reverse k-nearest neighbor

PurposeThis study aims to propose and evaluate a searching scheme for a bichromatic reverse k-nearest neighbor (BRkNN) that has objects and queries in spatial networks. In this proposed scheme, the author’s search for the BRkNN of the query using an influence zone for each object with a network Voronoi diagram (NVD).Design/methodology/approachThe author’s analyze and evaluate the performance of the proposed searching scheme.FindingsThe contribution of this paper is that it confirmed that the proposed searching scheme gives shorter processing time than the conventional linear search.Research limitations/implicationsA future direction of this study will involve making a searching scheme that reduces the processing time when objects move automatically on spatial networks.Practical implicationsIn BRkNN, consider two groups in a convenience store, where several convenience stores, which are constructed in Groups A and B, operate in a given region. The author’s can use RNN is RkNN when k = 1 (RNN) effectively to set a new store considering the Euclidean and road distances among stores and the location relationship between Groups A and B.Originality/valueIn the proposed searching scheme, the author’s search for the BRkNN of the query for each object with an NVD using the influence zone, which is the region where an object in the spatial network recognizes the nearest neighbor for the query.

On efficiently finding reverse k-nearest neighbors over uncertain graphs

Reverse k-nearest neighbor ($$\hbox {R}k\hbox {NN}$$RkNN) query on graphs returns the data objects that take a specified query object q as one of their k-nearest neighbors. It has significant influence in many real-life applications including resource allocation and profile-based marketing. However, to the best of our knowledge, there is little previous work on $$\hbox {R}k\hbox {NN}$$RkNN search over uncertain graph data, even though many complex networks such as traffic networks and protein---protein interaction networks are often modeled as uncertain graphs. In this paper, we systematically study the problem of reversek-nearest neighbor search on uncertain graphs ($$\hbox {UG-R}k\hbox {NN}$$UG-RkNN search for short), where graph edges contain uncertainty. First, to address $$\hbox {UG-R}k\hbox {NN}$$UG-RkNN search, we propose three effective heuristics, i.e., GSP, EGR, and PBP, which minimize the original large uncertain graph as a much smaller essential uncertain graph, cut down the number of possible graphs via the newly introduced graph conditional dominance relationship, and reduce the validation cost of data nodes in order to improve query efficiency. Then, we present an efficient algorithm, termed as SDP, to support $$\hbox {UG-R}k\hbox {NN}$$UG-RkNN retrieval by seamlessly integrating the three heuristics together. In view of the high complexity of $$\hbox {UG-R}k\hbox {NN}$$UG-RkNN search, we further present a novel algorithm called TripS, with the help of an adaptive stratified sampling technique. Extensive experiments using both real and synthetic graphs demonstrate the performance of our proposed algorithms.

Hub Labels on the database for large-scale graphs with the COLD framework

Shortest-path computation on graphs is one of the most well-studied problems in algorithmic theory. An aspect that has only recently attracted attention is the use of databases in combination with graph algorithms, so-called distance oracles, to compute shortest-path queries on large graphs. To this purpose, we propose a novel, efficient, pure-SQL framework for answering exact distance queries on large-scale graphs, implemented entirely on an open-source database engine. Our COLD framework (COmpressed Labels on the Database) can answer multiple distance queries (vertex-to-vertex, one-to-many, k-Nearest Neighbors, Reverse k-Nearest Neighbors, Reverse k-Farthest Neighbors and Top-k Range) not handled by previous methods, rendering it a complete database solution for a variety of practical large-scale graph applications. Our experimentation shows that COLD outperforms existing approaches (including popular graph databases) in terms of query time and efficiency, while requiring significantly less storage space than these methods.

Reverse Nearest Neighbor Search on a Protein-Protein Interaction Network to Infer Protein-Disease Associations.

The associations between proteins and diseases are crucial information for investigating pathological mechanisms. However, the number of known and reliable protein-disease associations is quite small. In this study, an analysis framework to infer associations between proteins and diseases was developed based on a large data set of a human protein-protein interaction network integrating an effective network search, namely, the reverse k-nearest neighbor (RkNN) search. The RkNN search was used to identify an impact of a protein on other proteins. Then, associations between proteins and diseases were inferred statistically. The method using the RkNN search yielded a much higher precision than a random selection, standard nearest neighbor search, or when applying the method to a random protein-protein interaction network. All protein-disease pair candidates were verified by a literature search. Supporting evidence for 596 pairs was identified. In addition, cluster analysis of these candidates revealed 10 promising groups of diseases to be further investigated experimentally. This method can be used to identify novel associations to better understand complex relationships between proteins and diseases.

Efficient Processing of Continuous Reverse k Nearest Neighbor on Moving Objects in Road Networks

A reverse k nearest neighbor (RkNN) query retrieves all the data points that have q as one of their k closest points. In recent years, considerable research has been conducted into monitoring reverse k nearest neighbor queries. In this paper, we study the problem of continuous reverse nearest neighbor queries where both the query object q and data objects are moving. Existing state-of-the-art techniques are sensitive towards the movement of data objects, e.g., a candidate object must be verified whenever it changes its location. Further, insufficient attention has been given to the monitoring of RNN queries in dynamic road networks where the network weight changes depending on the traffic conditions. In this paper, we address these problems by proposing a new safe exit-based algorithm called CORE-X for efficiently computing the safe exit points of both query and data objects. The safe exit point of an object indicates the point at which its safe region and non-safe region meet, thus a set of safe exit points represents the border of the safe region. Within the safe region, the query result remains unchanged provided the query and data objects remain inside their respective safe regions. The results of extensive experiments conducted using real road maps indicate that the proposed algorithm significantly reduces communication and computation costs compared to the state-of-the-art algorithm.

Reverse k nearest neighbors queries and spatial reverse top-k queries

Given a set of facilities and a set of users, a reverse k nearest neighbors (RkNN) query q returns every user for which the query facility is one of the k closest facilities. Almost all of the existing techniques to answer RkNN queries adopt a pruning-and-verification framework. Regions-based pruning and half-space pruning are the two most notable pruning strategies. The half-space-based approach prunes a larger area and is generally believed to be superior. Influenced by this perception, almost all existing RkNN algorithms utilize and improve the half-space pruning strategy. We observe the weaknesses and strengths of both strategies and discover that the regions-based pruning has certain strengths that have not been exploited in the past. Motivated by this, we present a new regions-based pruning algorithm called Slice that utilizes the strength of regions-based pruning and overcomes its limitations. We also study spatial reverse top-k (SRTk) queries that return every user u for which the query facility is one of the top-k facilities according to a given linear scoring function. We first extend half-space-based pruning to answer SRTk queries. Then, we propose a novel regions-based pruning algorithm following Slice framework to solve the problem. Our extensive experimental study on synthetic and real data sets demonstrates that Slice is significantly more efficient than all existing RkNN and SRTk algorithms.

ReHub

Quite recently, the algorithmic community has focused on solving multiple shortest-path query problems beyond simple vertex-to-vertex queries, especially in the context of road networks. Unfortunately, those advanced query-processing techniques cannot be applied to large-scale graphs, such as social or collaboration networks, or to efficiently answer reverse k -nearest neighbor (R k NN) queries, which are of practical relevance to a wide range of applications. To remedy this, we propose ReHub, a novel main-memory algorithm that extends the hub labeling technique to efficiently answer R k NN queries on large-scale networks. Our experimentation will show that ReHub is the best overall solution for this type of queries, requiring only minimal additional preprocessing and providing very fast query times in all cases.

Monochromatic and bichromatic ranked reverse boolean spatial keyword nearest neighbors search

Recently, Reverse k Nearest Neighbors (RkNN) queries, returning every answer for which the query is one of its k nearest neighbors, have been extensively studied on the database research community. But the RkNN query cannot retrieve spatio-textual objects which are described by their spatial location and a set of keywords. Therefore, researchers proposed a RSTkNN query to find these objects, taking both spatial and textual similarity into consideration. However, the RSTkNN query cannot control the size of answer set and to be sorted according to the degree of influence on the query. In this paper, we propose a new problem Ranked Reverse Boolean Spatial Keyword Nearest Neighbors query called Ranked-RBSKNN query, which considers both spatial similarity and textual relevance, and returns t answers with most degree of influence. We propose a separate index and a hybrid index to process such queries efficiently. Experimental results on different real-world and synthetic datasets show that our approaches achieve better performance.