Link prediction in paper citation network to construct paper correlation graph
 397 Downloads
 2 Citations
Abstract
Nowadays, recommender system has become one of the main tools to search for users’ interested papers. Since one paper often contains only a part of keywords that a user is interested in, recommender system returns a set of papers that satisfy the user’s need of keywords. Besides, to satisfy the users’ requirements of further research on a certain domain, the recommended papers must be correlated. However, each paper of an existing paper citation network hardly has cited relationships with others, so the correlated links among papers are very sparse. In addition, while a mass of research approaches have been put forward in terms of link prediction to address the network sparsity problems, these approaches have no relationship with the effect of selfcitations and the potential correlations among papers (i.e., these correlated relationships are not included in the paper citation network as their published time is close). Therefore, we propose a link prediction approach that combines time, keywords, and authors’ information and optimizes the existing paper citation network. Finally, a number of experiments are performed on the realworld HepTh datasets. The experimental results demonstrate the feasibility of our proposal and achieve good performance.
Keywords
Link prediction Paper citation network Paper correlated graph Time Keywords Authors’ informationAbbreviations
 AUC
Area under the curve
 KA
Keywords & Authors
 RAKE
Rapid Automatic Keyword Extraction algorithm
 ROC
Receiver operating characteristic
 TKA
Time & Keywords & Authors
 WCN
Weighted Common Neighbor
 WJC
Weighted Jaccard Coefficient
1 Introduction
Currently, users can type their preferred keywords into papersearching websites (e.g., Google Scholar and Baidu Academic) to search for their interested papers, and then, these websites will recommend appropriate papers to them [1]. Generally, a paper just contains partial keywords that a user is interested in, so a paper recommender system must return a set of papers that collectively cover all requested keywords.
As shown in Fig. 1, the user obtains a set of keywords including link prediction, weighting criteria, data mining, and citation network. Then, papersearching websites usually recommend some papers to their users based on those above keywords. As we all know, the keywords of a paper can only represent papers’ topics or themes; therefore, considering keywords only appear in papersearching process may find a set of papers that belong to different research domains or are actually not correlated, which fails to satisfy the user original requirements on deep and continuous research.
Fortunately, paper citation network that depicts the cited relationships among different papers has provided a promising way to model the correlations among the papers in terms of width and depth perspectives. However, the current paper citation network still faces a big challenge, that is, each paper of the existing paper citation network has slight cited relationships with other papers, so that correlated relationships among papers are also very sparse.
Considering this challenge, we will propose a novel link prediction approach to optimize the existing paper citation network. Furthermore, many previous researches proved that link prediction is the best solution to various network optimization problems [2, 3]. More specifically, link prediction attempts to estimate the likelihood of the existence of a link between two nodes because nodes attribute to information and network structures. In addition, when using our proposal to build new paper relationships (i.e., correlated relationships), we also consider the effect of selfcitations from authors and potential correlations among papers (i.e., these correlated relationships are not included in the paper citation network as their published time is close).

We propose a novel link prediction approach to construct new relation graphs. Our proposal considers a wide range of factors that influence the correlations among papers, such as paper published time, paper keywords, and paper authors. Furthermore, our link prediction approach takes the network structure of paper citation network into considerations, which makes the predicted results more reasonable and convincing.

We optimize the existing paper citation network by reducing the negative influence of intentional selfcitations from partial authors.

At last, extensive experiments are performed on a realworld paper dataset to demonstrate the actual capability of our method of dealing with the network sparsity problem.
The rest of paper is organized as follows. Related work is presented in Section 2. In Section 3, we introduce the research motivation. In Section 4, the detail of our proposed link prediction approach is described. Next, Section 5 discusses the experimental datasets (i.e., HepTh) and experimental evaluated metrics and mainly analyzes the experimental results. Finally, in Section 6, we have summarized our proposal as well as future research topics.
2 Related work
Link prediction is a significant research content and approach of optimizing various network. To the best of our knowledge, an essential fact of the link prediction is that node attributes to those known information and network structure features, so link prediction methods can easily find the missing links. Besides, these methods can build new links (i.e., correlated links) between two nodes without connection. Thus, the link prediction can effectively address a core problem of our proposal, i.e., solve the sparsity in the existing paper citation network.
Currently, link prediction has made massive strides and plays an important role in many research areas. For example, new friends through link prediction can be found in social network [4] and proteinprotein interactions can also be found [5]. Link prediction approaches can be classified into three categories: similaritybased methods, maximum likelihood approaches, and probabilistic methods [6]. As far as we know, the similaritybased methods can be used to the largescale networks, which is because it can calculate the similarity scores between two nodes [7]. Although maximum likelihood approaches can obtain specific parameters and probabilistic methods can predict missing links by using the trained model, maximum likelihood approaches and probabilistic methods cannot dispose of the broadscale networks [8]. Therefore, we mainly consider the similaritybased approach in our research.
Generally, the similaritybased approach can also be classified into two categories: the network structurebased similarity methods and the node attributebased similarity methods. The node attributebased similarity methods mainly focus on the node attribute to information of finding the similar nodes, so these methods are a significant way to form node pairs. Furthermore, these methods also solve the coldstart problem for link prediction research, e.g., Wang et al. [9] used the node attribute information (e.g., user profile) to address the coldstart problem on Twitter and Facebook. In addition, the network structurebased similarity method allocates similarity scores to the node pairs according to the structure features of networks. Currently, the network structurebased similarity method mainly contains four categories, i.e., local approaches, global approaches, quasilocal approaches, and communitybased approaches [10]. Here, we mainly pay attention to the local similaritybased approaches, because it calculates the similarity scores of two nodes without connection based on the nodes’ neighboring structural features; furthermore, some common index of the local approaches can be used in the largescale networks, e.g., Common Neighbors index (CN), Jaccard Coefficient (JC), Adamic–Adar index (AA), and Resource Allocation index (RA).
Many of link prediction researches only concentrate on unweighted networks, but actually, many realworld networks can be weighted. For example, edge weighting value can represent the strength of connection in brain networks and the number of flights in airline networks [11], respectively. For the social network, the work [12] uses local weighted similarity functions to calculate the weighting value of two nodes without connection. Besides, this work also proves that the weak ties have an effect on link prediction. In addition, [13] shows that the ties of spouses or romantic partners play an important role in the social network, so these ties can be regarded as one of the significant edgeweighted ways in the link prediction. Recently, the work in [14] is carrying out a study into the effects of the strength of link in the social network and proposes weighting criterion for link prediction model according to users’ data information and the number of interactions among users. However, in their weighting criterion, their work does not take full advantage of the node and its attribute information.
In view of the above research content, we know that the link prediction is one of the significant approaches to solve network sparsity, as it specializes in predicting the missing/correlated links among two nodes without connection. Thus, we propose a novel link prediction approach to construct the paper correlated graph, that is, the similaritybased weighting method.
3 Research motivation
In our paper, we focus on the following key issue: how to solve the sparsity of the existing paper citation network? As for this problem, link prediction approach is the best solution. Furthermore, in the process of building a correlated relationship on the paper citation network, we consider the effect of selfcitations from authors and potential correlations among the papers, which are not included in citation network but with close published time.
4 Link prediction method

Activity 1: Preprocessing of the network. In order to construct a paper correlated graph, the paper citation network is regarded as an undirected paper citation network (G).

Activity 2: To divide paper citation network. G is partitioned into two parts, i.e., G_{train} and G_{test}. In the G_{train}, we need to get the average score from existing pairs of nodes. Furthermore, in the G_{test}, we need to get the weighting value of the two nodes without connection.

Activity 3: Network to be weighted. In the G_{train}, the weighting value of the two connected nodes are calculated by using the weighting criteria, and the weighting value of two nodes without connection are calculated in the G_{test}.

Activity 4: Score calculation and ranking. (1) Firstly, we use two weighted similarity function formulas WCN and WJC [16] to calculate the weighting value of two nodes without connection in the G_{train}. Next, we can obtain a ranking list in order to descend weighting value. At last, the average score is saved in w_{average}(v_{itrain} , v_{jtrain}).
 (A)
Weighted Common Neighbor  WCN(v_{itrain}, v_{jtrain}). Which is:
 (B)
Weighted Jaccard Coefficient  WJC(v_{itrain}, v_{jtrain}). Which is:
Equations (5) and (6) are used to find the maximum value and the minimum value in the G_{train}.

Activity 5: Connected nodes. LP (link prediction) is defined as in Eq. (8):
4.1 Proposed weighting criteria
Here, we propose weighting model which can generate two disparate weighting criteria based on the Eq. (9). Please note that the product between paper’s data in weighting criteria emphasizes the fact that the selected data information must be concurrently calculated. Thus, the two disparate weighting criteria are as follows:
4.1.1 Keywords and authors’ weighting criteria
where α and β (0 < α, β < 1) are arbitrary damping parameters and they are used to calibrate the importance of paper authors and paper keywords in the weighting criteria. \( {A}_{v_i}^a \) (\( {A}_{v_j}^a \)) and \( {K}_{v_i} \) (\( {K}_{v_j} \)) are a set of authors and keywords, respectively. \( {A}_{v_i}^a\cap {A}_{v_j}^a \) and \( {K}_{v_i}\cap {K}_{v_j} \) represent the coauthors and common keywords, respectively. \( \mathrm{cosine}\left({A}_{v_i}^a,{A}_{v_j}^a\right) \) and \( \mathrm{cosine}\left({K}_{v_i},{K}_{v_j}\right) \) denote an approach that computes the similarity between the two nodes v_{i} and v_{j}, respectively. A constant C is defined for convenience of calculation.
4.1.2 Time, keywords, and authors’ weighting criteria
where parameter λ (0 < λ < 1) adjusts the effect of paper time on the weighting criteria. Furthermore, \( {t}_{v_i} \) and \( {t}_{v_j} \) indicate the published time of nodes v_{i} and v_{j}.
Note that, if there are no common keywords and coauthors among two papers, the weighting value of them will be set to the fixed value, namely w_{n − co} = 0.1. In addition, as for synonymy, word inflections and polysemy are tackled with automatic query expansion techniques [17]. However, it is out of the scope of this paper research.
4.2 Paper correlated graph
Here, we define the undirected relation network as a paper correlated graph.
Definition 1. Paper correlated graph: Paper correlated graph is represented by G_{p} = {V_{p}, E_{p}}, where V_{p} and E_{p} denote its set of nodes and edges, respectively. Furthermore, for each of node pairs (v_{i}, v_{j}), the paper correlated graph has a corresponding edge e(v_{i}, v_{j}).
5 Experiments
In this section, largescale experiments are designed and tested on a realworld paper citation dataset which demonstrates the usefulness and effectiveness of our link prediction approach.
5.1 Experimental environments
5.1.1 Experimental tools
The proposed link prediction approach is implemented in PyCharm and executed under the environment of Intel(R) Core(R) CPU @3.0 GHz, 16 GB RAM, and Windows 10 @ 1809, 64bit operating system.
5.1.2 Dataset
A paper citation network was extracted from the available HepTh dataset [18]. In the paper citation network, a node represents a paper and an edge indicates that two specific nodes have cited relationship. Furthermore, each node stores the paper’s published time, keywords, and authors’ information. Besides, we will use the information of each paper that the title and abstract are used to construct a set of keywords; here, we mainly use RAKE (Rapid Automatic Keyword Extraction algorithm) method to construct a set of keywords, which is because this method analyzes the frequency of words appearing and their cooccurrence with other words is used to identify keywords or phrases in the body of a text.
Our experimental process of link prediction also follows the task sequence that is depicted in Fig. 3. First, the existing paper citation network is partitioned into two parts according to their published time. So, the existing paper citation network from the HepTh dataset is partitioned into G_{train} = [1997, 1999] and G_{test} = [2000, 2002]. The G_{train} contains 7304 nodes (papers) and 56,376 edges; likewise, the G_{test} contains 8721 nodes (papers) and 70,045 edges. Next, we need to configure parameters in our experiment (i.e., α, β, and λ). In this experimental process, we need to finetune each parameter of two different weighting criteria to find the accurate and credible parameter values. Thus, for parameter values α, β, and λ, we first range the values of α from 0.3 to 0.9 with step 0.2, we range the values of β from 0.3 to 0.9 with step 0.1; and finally, we set the values of λ to 0.3 and 0.9 to reflect the factors of published time. In addition, in order to calculate a value of two nodes without connection in the G_{train}, we select two disparate weighted similarity functions, i.e., WCN and WJC, as well as these two disparate weighted similarity functions are usually used in different link prediction research. Next, these two weighted similarity functions will be combined with two different weighting criteria (i.e., Keywords & Authors (KA) and Time & Keywords & Authors (TKA)). Thus, we will obtain four disparate functions (i.e., WCN^{KA}, WCN^{TKA}, WJC^{KA}, WCN^{TKA}) and apply these functions into our experiments.
5.1.3 Evaluation criteria
 (1)
AUC (area under the receiver operating characteristic curve [19]). The area under the ROC (receiver operating characteristic) curve can demonstrate link prediction methods accuracy, so the AUC can be regarded as measure index. In our research, we first assign a weighting value for each correlated links and nonexistent links in the paper citation network. Next, we will randomly select correlated links and nonexistent links and compare their values. Finally, we can obtain AUC value by acting the Eq. (16).
 (2)
Edge number. For link prediction approach, we argue that the newer generated edges it has, the approach will be better. Therefore, we test the number of the new generated edges of the output graph (i.e., the paper correlated graph).
 1)
Random [20]: On the test network, if two nodes without connection have more than half of the number of same keywords, these two nodes without connection will generate new edges.
 2)
Maximum [20]: If the weighting value of two nodes without connection in the test network is greater than the maximum value found in the training network, these two nodes without connection will generate new edges.
5.2 Experimental results
5.2.1 Profile 1: the AUC value of five approaches
As shown in Figs. 4, 5, 6, and 7, for our proposal (i.e., OurKA and OurTKA) and the maximum approach (i.e., MaximumKA and MaximumTKA), with the same weighted similarity functions, the value of AUC generally increases as the parameters value increases. That is because the node attribute information plays an increasingly important effect on link prediction during the process of parameter value increasing. However, for the random approach, the value of AUC, it is not affected by the parameters value and the weighted similarity functions. In addition, Figs. 4, 5, 6, and 7 show that, with the same weighted similarity functions and the parameters value, the AUC values obtained by our proposal are generally greater than those obtained by the maximum and random approaches. Further, it shows that our proposal is superior to other approaches. As far as we know, the larger value of AUC means that our proposal can better improve the existing paper citation network, i.e., our proposal played a significant role in lightening the sparsity of the existing paper citation network.
5.2.2 Profile 2: the number of new edges built by five approaches
The different parameters are employed in the weighted similarity function WCN, and λ = 0.3 to obtain the edge number values
Approaches  α/β/edge number  

0.3  0.5  0.7  0.9  
Random  4  4  4  4 
MaximumKA  12  12  12  10 
MaximumTKA  16  14  14  14 
OurKA  647  4640  26,342,452  26,344,476 
OurTKA  380  2606  8,802,878  21,618,518 
The different parameters are employed in the weighted similarity function WCN, and λ = 0.9 to obtain the edge number values
Approaches  α/β/edge number  

0.3  0.5  0.7  0.9  
Random  4  4  4  4 
MaximumKA  12  12  12  10 
MaximumTKA  18  18  18  16 
OurKA  647  4640  26,342,452  26,344,476 
OurTKA  638  4588  12,339,824  26,344,484 
The different parameters are employed in the weighted similarity function WJC, and λ = 0.3 to obtain the edge number values
Approaches  α/β/edge number  

0.3  0.5  0.7  0.9  
Random  4  4  4  4 
MaximumKA  128  4122  26,342,452  26,344,476 
MaximumTKA  6  8  14  740 
OurKA  5384  26,342,452  26,344,478  26,344,476 
OurTKA  182  2074  8,818,608  26,344,484 
The different parameters are employed in the weighted similarity function WJC, and λ = 0.9 to obtain the edge number values
Approaches  α/β/edge number  

0.3  0.5  0.7  0.9  
Random  4  4  4  4 
MaximumKA  128  4122  26,342,452  26,344,476 
MaximumTKA  6  8  14  740 
OurKA  5384  26,342,452  26,344,478  26,344,476 
OurTKA  4534  12,340,892  26,344,484  26,344,484 
As shown in Tables 1, 2, 3, and 4, for our proposal (i.e., OurKA and OurTKA) and the maximum approach (i.e., MaximumKA and MaximumTKA), under the same weighted similarity functions, the number of new edges would increase as the parameter value increases. That is because the node attribute information plays an increasingly important role in link prediction as the parameter value increases. Furthermore, the new edges built by our approach are larger than the maximum and random approaches, which show that our proposal is superior to other approaches. In addition, for our proposal, with the same parameter value and the same weighting criteria (i.e., KA and TKA), we find that the number of the new edges built by the weighted similarity function WJC is generally greater than that built by WCN. It shows that the WJC achieves better results in link prediction than the WCN. According to the above analysis, our proposal can effectively solve sparsity of the existing paper citation network.
5.2.3 Profile 3: investigate the performance in different parameter value setting and the weighted similarity functions
5.2.4 Profile 4: performance comparison in different functions
The result of performance metrics in the different functions
Similarity function  Performance  

AUC  Edge number  
WCN^{KA}  0.9992  26,344,478 
WCN^{TKA}  0.9993  26,344,484 
WJC^{KA}  0.9992  26,344,478 
WJC^{TKA}  0.9993  26,344,484 
5.3 Further discussions
However, there are still some shortages in our proposed link prediction approach. Firstly, there are many attributes of node on the existing paper citation network, and it is hard to obtain these attribute information [21, 22] to further optimize paper citation network. Secondly, since the process of obtaining data may involve some privacy issues [23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33], our work will further consider the privacypreservation effects when making link prediction. Finally, more complex multidimensional or multicriterion application scenarios [34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44] should be considered in the future to make our proposal more comprehensive.
6 Conclusions
Predicting whether two correlated papers will build correlated links in an existing paper citation network is a significant analysis task, which is regarded as a link prediction problem. To find and build correlated links in the existing paper citation network, we put forward a novel link prediction approach. The novel link prediction approach not only has advantages of predicting and building correlated links, but also helps with alleviating the current paper citation network sparsity. Furthermore, we also use the combination of paper time, paper keywords, and paper authors’ information to reduce the effect of the selfcitations. Since the weighting value of nodes pair in the paper citation network is obtained from calculating its attribute information, the experimental results can reflect the actual weighting value of a node pair as accurately as possible. Finally, the feasibility of our proposal is validated by a realworld dataset.
In the future, we will continue to refine our work by considering more complex scenarios, such as privacyaware or multidimensional link prediction problems.
Notes
Acknowledgements
Not applicable.
Authors’ contributions
HL finished the algorithm and English writing of the paper. HK and CY finished the experiments. LQ put forward the idea of this paper. All authors read and approved the final manuscript.
Funding
This work was supported by the National Key Research and Development Program of China (No. 2017YFB1400600) and the Natural Science Foundation of China (No. 61872219).
Competing interests
The authors declare that they have no competing interests.
References
 1.L. Pan et al., Academic Paper Recommendation Based on Heterogeneous Graph. Chinese Computational Linguistics and Natural Language Processing Based on Naturally Annotated Big Data (Springer, Guangzhou, 2015), pp. 381–392Google Scholar
 2.X. Zhou et al., Academic influence aware and multidimensional network analysis for research collaboration navigation based on scholarly big data. IEEE Trans. Emerg. Top. Comput. (2018). https://doi.org/10.1109/TETC.2018.2860051
 3.X. Zhou et al., Analysis of user network and correlation for community discovery based on topicaware similarity and behavioral influence. IEEE Trans. Hum. Mach. Syst. 48(6), 559–571 (2018). https://doi.org/10.1109/THMS.2017.2725341 MathSciNetCrossRefGoogle Scholar
 4.L.M. Aiello et al., Friendship prediction and homophily in social media. ACM Trans. Web (TWEB) 6(2), 9 (2012)Google Scholar
 5.C. Lee et al., How to assess patent infringement risks: A semantic patent claim analysis using dependency relationships. Tech. Anal. Strat. Manag. 25, 23–38 (2013)CrossRefGoogle Scholar
 6.L. Lü et al., Link prediction in complex networks: A survey. Phys. A 390(6), 1150–1170 (2011)CrossRefGoogle Scholar
 7.A. Clauset et al., Hierarchical structure and the prediction of missing links in networks. Nat. Publ. Group 453, 98–101 (2008)Google Scholar
 8.R.H. Li et al., in Proceedings of the 20th ACM International Conference on Information and Knowledge Management. CIKM’11. Link prediction: The power of maximal entropy random walk (2011), pp. 1147–1156Google Scholar
 9.Z. Wang et al., An approach to coldstart link prediction: Establishing connections between nontopological and topological information. IEEE Trans. Knowl. Data Eng. 28(11), 2857–2870 (2016)CrossRefGoogle Scholar
 10.E. Bastami et al., A gravitationbased link prediction approach in social networks. Swarm Evol. Comput. Available online (2018)Google Scholar
 11.M.E.J. Newman et al., The structure and function of complex networks. SIAM Rev. 45(2), 167–256 (2003)MathSciNetCrossRefGoogle Scholar
 12.L. Lü et al., Link prediction in weighted networks: The role of weak ties. Europhys. Lett. Assoc. 89(1), 18001 (2010)CrossRefGoogle Scholar
 13.L. Backstrom et al., in Proc. 17th ACM Conf. Comput. Supported Cooperative Work Social Comput. Romantic partnerships and the dispersion of social ties: A network analysis of relationship status on Facebook (2014), pp. 831–841Google Scholar
 14.C.P. Muniz et al., Combining contextual, temporal and topological information for unsupervised link prediction in social network. Knowl.Based Syst. 156, 129–137 (2018)CrossRefGoogle Scholar
 15.D. LibenNowell et al., The linkprediction problem for social networks. J. Am. Soc. Inf. Sci. Technol. 58(7), 1019–1031 (2007)CrossRefGoogle Scholar
 16.P.M. Chuan et al., Link prediction in coauthorship networks based on hybrid content similarity metric. Appl. Intell. 48(8), 2470–2248 (2018)CrossRefGoogle Scholar
 17.L. Qi et al., TimeAware IoE Service Recommendation on Sparse Data. Mob. Inf. Syst. 2016, 4397061, 12 Pages (2016)Google Scholar
 18.J. Leskovec et al., in ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD). Graphs over Time: Densification Laws, Shrinking Diameters and Possible Explanations (2005)Google Scholar
 19.Z. Wen et al., hindexbased link prediction methods in citation network. Scientometrics (2018)Google Scholar
 20.L. Qi et al., Finding all you need: Web APIs recommendation in web of things through keywords search. IEEE Trans. Comput. Soc. Syst. (2019). https://doi.org/10.1109/TCSS.2019.2906925
 21.W. Li et al., On improving the accuracy with autoencoder on conjunctivitis. Appl. Soft Comput. 81, 105489 (2019)CrossRefGoogle Scholar
 22.S. Ding et al., Image caption generation with highlevel image features. Pattern Recogn. Lett. 123, 89–95 (2019)CrossRefGoogle Scholar
 23.Z. Gao et al., Adaptive fusion and categorylevel dictionary learning model for multiview human action recognition. IEEE Internet Things J. (2019)Google Scholar
 24.L. Qi et al., Timeaware distributed service recommendation with privacypreservation. Inf. Sci. 480, 354–364 (2019)CrossRefGoogle Scholar
 25.S. Zhang et al., A caching and spatial Kanonymity driven privacy enhancement scheme in continuous locationbased services. Futur. Gener. Comput. Syst. 94, 40–50 (2019)CrossRefGoogle Scholar
 26.X. Wang et al., Improved multiorder distributed HOSVD with its incremental computing for smart city services. IEEE Trans. Sustain. Comput. (2018). https://doi.org/10.1109/TSUSC.2018.2881439
 27.W. Gong et al., Privacyaware multidimensional mobile service quality prediction and recommendation in distributed fog environment. Wirel. Commun. Mob. Comput. 2018, 3075849, 8 pages (2018)Google Scholar
 28.S. Zhang et al., A trajectory privacypreserving scheme based on a dualK mechanism for continuous locationbased services. Inf. Sci. (2019). https://doi.org/10.1016/j.ins.2019.05.054
 29.X. Xu et al., An edge computingenabled computation offloading method with privacy preservation for internet of connected vehicles. Futur. Gener. Comput. Syst. 96, 89–100 (2019)CrossRefGoogle Scholar
 30.L. Qi et al., A distributed localitysensitive hashing based approach for cloud service recommendation from multisource data. IEEE J. Sel. Areas Commun. 35(11), 2616–2624 (2017)CrossRefGoogle Scholar
 31.Y. Xu et al., Privacypreserving and scalable service recommendation based on SimHash in a distributed cloud environment. Complexity 2017, 3437854, 9 pages (2017)Google Scholar
 32.X. Wang et al., A cloudedge computing framework for cyberphysicalsocial services. IEEE Commun. Mag. 55(11), 80–85 (2017)CrossRefGoogle Scholar
 33.L. Qi et al., A twostage localitysensitive hashing based approach for privacypreserving mobile service recommendation in crossplatform edge environment. Futur. Gener. Comput. Syst. 88, 636–643 (2018)CrossRefGoogle Scholar
 34.Y. Zhang et al., Service recommendation based on quotient space granularity analysis and covering algorithm on Spark. Knowl.Based Syst. 147, 25–35 (2018)CrossRefGoogle Scholar
 35.Z. Gao et al., Cognitiveinspired classstatistic matching with tripleconstrain for camera free 3D object retrieval. Futur. Gener. Comput. Syst. 94, 641–653 (2019)CrossRefGoogle Scholar
 36.S. Wan et al., Multidimensional data indexing and range query processing via Voronoi diagram for internet of things. Futur. Gener. Comput. Syst. 91, 382–391 (2019)CrossRefGoogle Scholar
 37.X. Wang et al., A tensorbased big datadriven routing recommendation approach for heterogeneous networks. IEEE Netw. Mag. 33(1), 64–69 (2019)CrossRefGoogle Scholar
 38.S. Ding et al., A long video caption generation algorithm for big video data retrieval. Futur. Gener. Comput. Syst. 93, 583–595 (2019)CrossRefGoogle Scholar
 39.L. Qi et al., A QoSaware virtual machine scheduling method for energy conservation in Cloudbased CyberPhysical Systems. World Wide Web J. (2019). https://doi.org/10.1007/s1128001900684y
 40.S. Zhang et al., Enhancing privacy through uniform grid and caching in locationbased services. Futur. Gener. Comput. Syst. 86, 881–892 (2018)CrossRefGoogle Scholar
 41.L. Qi et al., Dynamic mobile crowdsourcing selection for electricity load forecasting. IEEE Access 6, 46926–46937 (2018)CrossRefGoogle Scholar
 42.Y. Zhang et al., Coveringbased web service quality prediction via neighborhoodaware matrix factorization. IEEE Trans. Serv. Comput. (2019). https://doi.org/10.1109/TSC.2019.2891517
 43.S. Zhang et al., A dual privacy preserving scheme in continuous locationbased services. IEEE Internet Things J. 5(5), 4191–4200 (2018)CrossRefGoogle Scholar
 44.X. Wang et al., NQA: A nested anticollision algorithm for RFID systems. ACM Trans. Embed. Comput. Syst. 18(4) (2019). https://doi.org/10.1145/3330139
Copyright information
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.