Personalized PageRank Algorithm Research Articles

A Subgraph Retrieval Method for Complex Questions Based on Hybrid Semantics and Path Representation

Current subgraph retrieval methods generally fall into two categories: those that rely on semantic matching, which use only surface-level semantic information of relations and lack flexibility; and those based on personalized PageRank algorithms, which fail to leverage the semantic connection between the relation and the question, rendering them susceptible to noisy data. To address these issues, this paper introduces a novel retrieval model that employs hybrid semantics of relations and path representations. Specifically, hybrid semantics involves merging relational and entity information within a knowledge graph to extract the deep semantics of relations and enhance semantic representation by integrating it with the explicit descriptive text of the relations. Path representation merges the semantics of the current relation with those of preceding ones to form a complete path representation. This representation is then semantically matched with the question to compute a score, which determines whether the relation should form part of the subgraph. We integrated our subgraph retrieval model with the Neural Symbolic Machine (NSM) reasoning model and evaluated it on the publicly available CWQ and WebQSP datasets. The experimental results demonstrate that our method performs exceptionally well on these datasets, validating the efficacy of utilizing deep semantics and path representations for the retrieval of subgraphs in response to complex questions.

GPPR: Geo-distributed Personalized PageRank Algorithm

GPPR: Geo-distributed Personalized PageRank Algorithm

Overlapping community detection in software ecosystem based on pheromone guided personalized PageRank algorithm

Software ecosystem has aroused the interest of numerous researchers and plays an important role in many aspects. According to different participants and their relationships, software ecosystem can be constructed into various types of complex networks. Overlapping community detection in complex networks can help reveal the community structure and find the intersection between communities. However, existing overlapping community detection algorithms often suffer from reduced applicability or accuracy when applied to networks in software ecosystems. To reveal the overlapping community structure in software ecosystem, we propose an overlapping community detection algorithm by improving the standard personalized PageRank (PPR) algorithm. We first construct a developer collaboration network in software ecosystem based on the intensity of cooperation between developers. Then, the similarity between developers is calculated to guide the walking process of the PPR algorithm, making it suitable for weighted networks. Finally, in the proposed algorithm PGPPR, inspired by the idea of using pheromones to guide the walking in ant colony algorithm, we run the algorithm in multiple rounds and use the results of the previous round to guide the walking process in the current round to reduce redundant diffusion. The experimental results on the five real-world networks show that our algorithm is applicable and effective in detecting communities. And in the five developer collaboration networks, PGPPR can effectively detect overlapping community structures with higher stability and accuracy than the four baselines. The PGPPR algorithm can find overlapping community structures in weighted networks and effectively reduce the redundant diffusion generated when applying the standard PPR algorithm to community detection. Compared to other algorithms, our algorithm can detect the overlapping communities more accurately and stably when applied to developer collaboration networks in software ecosystem.

A New BAT and PageRank Algorithm for Propagation Probability in Social Networks

Social networks have increasingly become important and popular in modern times. Moreover, the influence of social networks plays a vital role in various organizations, including government organizations, academic research organizations and corporate organizations. Therefore, strategizing the optimal propagation strategy in social networks has also become more important. Increasing the precision of evaluating the propagation probability of social networks can indirectly influence the investment of cost, manpower and time for information propagation to achieve the best return. This study proposes a new algorithm, which includes a scale-free network, Barabási–Albert model, binary-addition tree (BAT) algorithm, PageRank algorithm, Personalized PageRank algorithm and a new BAT algorithm to calculate the propagation probability of social networks. The results obtained after implementing the simulation experiment of social network models show that the studied model and the proposed algorithm provide an effective method to increase the efficiency of information propagation in social networks. In this way, the maximum propagation efficiency is achieved with the minimum investment.

Gauging node consistency in accusation–endorsement networks

Abstract Many signed, directed social networks can be viewed as being composed of positive (endorsements) and negative (accusations) directed edges, and these networks can in turn be created through a variety of different processes. The recently proposed consistency dynamics supposes that when nodes expect to be judged based on their associations in the network, they may create edges out of a desire to appear as having consistent judgements. We develop a quantifiable score that can rate the level of consistency in a node’s judgement. We demonstrate that this consistency score can be efficiently estimated using a modification of the popular personalized PageRank algorithm and evaluate the score’s properties. In order to validate this score’s relevance to empirical networks, we use consistency scores to perform an edge prediction task, and demonstrate that it performs competitively with, and adds complementary information to, more complicated measures designed specifically for that task. We also demonstrate that the nodes in these networks exhibit specific behaviours that consistency can identify across a range of parameterization values and which are not recoverable by other measures in isolation.

A local updating algorithm for personalized PageRank via Chebyshev polynomials

The personalized PageRank algorithm is one of the most versatile tools for the analysis of networks. In spite of its ubiquity, maintaining personalized PageRank vectors when the underlying network constantly evolves is still a challenging task. To address this limitation, this work proposes a novel distributed algorithm to locally update personalized PageRank vectors when the graph topology changes. The proposed algorithm is based on the use of Chebyshev polynomials and a novel update equation that encompasses a large family of PageRank-based methods. In particular, the algorithm has the following advantages: (i) it has faster convergence speed than state-of-the-art alternatives for local personalized PageRank updating; and (ii) it can update the solution of recent extensions of personalized PageRank that rely on complex dynamical processes for which no updating algorithms have been developed. Experiments in a real-world temporal network of an autonomous system validate the effectiveness of the proposed algorithm.

Conversational Recommender Systems Based on Mobile Chatbot for Culinary

Culinary places are one of the tourists attractions in a place that makes many new culinary places appear. Various types of new foods and drinks are present along with the addition of culinary places. However, this can be a problem when tourists visit a new destination and look for a culinary place that suits their tastes. In the previous research on the recommendation system for culinary places, users only gave their preferences at the beginning of the recommendation process and ignored the operating hours of the recommended culinary places. Therefore, we developed a recommendation system for culinary places by utilizing the context of time from users. We use the Conversational Recommender System on the chatbot platform with the Personalized PageRank algorithm to generate recommendations. In addition, we also use the explanation facility to get an explanation of the recommended items. We use questionnaires and the accuracy of recommendation results to measure user satisfaction and system performance. The evaluation results with a questionnaire involving 81 respondents concluded that users are pretty satisfied with the system built. However, testing with accuracy yields a value of 40%, proving that the system performance is low

Multilayer network analysis for improved credit risk prediction

We present a multilayer network model for credit risk assessment. Our model accounts for multiple connections between borrowers (such as their geographic location and their economic activity) and allows for explicitly modelling the interaction between connected borrowers. We develop a multilayer personalized PageRank algorithm that allows quantifying the strength of the default exposure of any borrower in the network. We test our methodology in an agricultural lending framework, where it has been suspected for a long time default correlates between borrowers when they are subject to the same structural risks. Our results show there are significant predictive gains just by including centrality multilayer network information in the model, and these gains are increased by more complex information such as the multilayer PageRank variables. The results suggest default risk is highest when an individual is connected to many defaulters, but this risk is mitigated by the size of the neighbourhood of the individual, showing both default risk and financial stability propagate throughout the network.

BATON: Batch One-Hop Personalized PageRanks with Efficiency and Accuracy

Personalized PageRank (PPR) is a classic measure of the relevance among different nodes in a graph, and has been applied in numerous systems, such as Twitter's Who-To-Follow and Pinterest's Related Pins. Existing work on PPR has mainly focused on three general types of queries, namely, single-pair PPR, single-source PPR, and all-pair PPR. However, we observe that there are applications that rely on a new query type (referred to as batch one-hop PPR ), which takes as input a set $S$ S of source nodes and, for each node $s \in S$ s ∈ S and each of $s$ s 's neighbor $v$ v , asks for the PPR value of $v$ v with respect to $s$ s . None of the existing PPR algorithms is able to efficiently process batch one-hop queries, due to the inherent differences between batch one-hop PPR and the three general query types. To address the limitations of existing algorithms, this paper presents Baton , an algorithm for batch one-hop PPR that offers both strong theoretical guarantees and practical efficiency. Baton leverages the characteristics of one-hop PPR to avoid unnecessary computation, and it incorporates advanced mechanisms to improve the cost-effectiveness of PPR derivations. Extensive experiments on benchmark datasets show that Baton is up to three orders of magnitude faster than the state of the art, while offering the same accuracy.

Linking chemical and disease entities to ontologies by integrating PageRank with extracted relations from literature

BackgroundNamed Entity Linking systems are a powerful aid to the manual curation of digital libraries, which is getting increasingly costly and inefficient due to the information overload. Models based on the Personalized PageRank (PPR) algorithm are one of the state-of-the-art approaches, but these have low performance when the disambiguation graphs are sparse.FindingsThis work proposes a Named Entity Linking framework designated by Relation Extraction for Entity Linking (REEL) that uses automatically extracted relations to overcome this limitation. Our method builds a disambiguation graph, where the nodes are the ontology candidates for the entities and the edges are added according to the relations established in the text, which the method extracts automatically. The PPR algorithm and the information content of each ontology are then applied to choose the candidate for each entity that maximises the coherence of the disambiguation graph. We evaluated the method on three gold standards: the subset of the CRAFT corpus with ChEBI annotations (CRAFT-ChEBI), the subset of the BC5CDR corpus with disease annotations from the MEDIC vocabulary (BC5CDR-Diseases) and the subset with chemical annotations from the CTD-Chemical vocabulary (BC5CDR-Chemicals). The F1-Score achieved by REEL was 85.8%, 80.9% and 90.3% in these gold standards, respectively, outperforming baseline approaches.ConclusionsWe demonstrated that RE tools can improve Named Entity Linking by capturing semantic information expressed in text missing in Knowledge Bases and use it to improve the disambiguation graph of Named Entity Linking models. REEL can be adapted to any text mining pipeline and potentially to any domain, as long as there is an ontology or other knowledge Base available.

WordNet construction for under-resourced languages using personalized PageRank

Abstract Wordnet development is an active research area among NLP researchers. Since the manual construction of the English wordnet was very costly both in terms of time and human expertise, automatic approaches have become very popular for wordnet development in languages other than English. Automatic methods usually benefit from an existing wordnet of a high resource language and use it as the backbone of their work. In this article, we present an unsupervised approach for automatic wordnet construction using a combination of Expectation–Maximization and personalized PageRank algorithms. Our method uses some typical and available language resources, so it is applicable to many languages including under-resourced ones. The proposed method needs just a bilingual dictionary and a monolingual corpus for developing a wordnet. In order to evaluate the proposed method, we apply it to the Persian language which is identified as an under-resourced language in NLP tasks. Evaluation results properly indicate the power of the proposed method to construct a high quality and large-scale wordnet for poor-resource languages. According to experiments, we achieve a precision of higher than 93% with a recall of 50%.

A novel approach for ranking web documents based on query-optimized personalized pagerank

Ranking plays an important role in the search process of web documents on a huge corpus. This not only reduces the searching time but also provides useful documents to the users. In this paper, we extend our earlier query-optimized PageRank approach by combining the TF-IDF and personalized PageRank algorithm to generate a robust ranking mechanism. In our earlier approach, we modeled a ranking scheme by considering the link structures of the documents along with their content. A novel feature selection technique named as ‘Term-term correlation-based feature selection’ (TCFS) is also proposed which removes all noise terms from the document before the ranking process starts. We believe that by incorporating TCFS and personalized PageRank of the documents along with their relevance will improve the retrieval results. The aim is to modify the link structure based on the similarity score between the content of the document and the user query. Experimental results show that the proposed feature selection technique can outperform the conventional feature selection techniques, and the performance of the combined TF-IDF and personalized PageRank approach is promising compared to the traditional approaches.

RETRACTED ARTICLE: Real-time personalization and recommendation in Adaptive Learning Management System

Various e-learning environments have been developed to provide sufficient materials for learners and thereby guide them to gain knowledge in any domain. Even though there were several factors that determine the motivation of the student, the skill set possessed by them is definitely an important factor. Hence in our proposed work, the behavioural and educational skill of the learners is tested by a skill test and learning content is provided only based on their skill test evaluation reports. The entire process is done by real time personalization based Adaptive Learning Management System and Personalized Page Rank algorithm. Finally, the Navies’ Bayes classifier was employed to classify the learners based on the performance over the skill test. Learners on their own can express both the optimistic and adverse skills that considerably impact the adaptive learning in e-learning environment. High skilled categorized learners are offered with advanced, intermediate learners with moderate and beginner or slow learners are provided with basic level of study content. The results are analysed based on their skills, individual's method of learning and time constraints.

SPread: Exploiting Fractal Social Community for Efficient Multi-Copy Routing in Taxi VDTNs

In this paper, we focus on the multi-copy routing in Vehicle Delay Tolerant Networks (VDTNs) consisted of taxis. Multi-copy routing can balance the network congestion caused by broadcasting and the efficiency limitation in single-copy routing. However, current multi-copy routing algorithms often let different copies of each packet search the destination node independently, which leads to a low utilization of copies. To solve this problem, we propose a fractal Social community based efficient multi-coPy routing algorithm in VDTNs, namely SPread, by taking advantage of social network features (i.e., weak ties and fractal community structure) measured in two large taxi-based Vehicle NETwork (VNET) traces. SPread carefully scatters different copies of each packet to different communities towards the destination community. This ensures that different copies search the destination community through different weak ties. For the routing of each copy, current routing algorithms either fail to exploit nodes’ reachability information to others (e.g., centrality based methods) or only use single-hop reachability information (e.g., community based methods). In order to overcome such drawbacks, inspired by the personalized PageRank algorithm, we design new algorithms for calculating vehicles’ multi-hop reachability to different communities/vehicles dynamically. Therefore, the routing efficiency of each copy can be enhanced. Furthermore, we propose an Advanced SPread (denoted by ASPread) by exploiting the spatio-contact correlation of the community and the different sizes of communities. ASPread also sends a larger number of copies of a packet to a community with a larger number of vehicles and vice versa. Finally, extensive trace-driven simulation demonstrates the high efficiency of SPread in comparison with state-of-the-art routing algorithms in DTNs, as well as the enhancement brought by ASPread.

Realtime top-k personalized pagerank over large graphs on GPUs

Given a graph G , a source node s ∈ G and a positive integer k , a top- k Personalized PageRank (PPR) query returns the k nodes with the highest PPR values with respect to s , where the PPR of a node v measures its relevance from the perspective of source s. Top- k PPR processing is a fundamental task in many important applications such as web search, social networks, and graph analytics. This paper aims to answer such a query in realtime , i.e., within less than 100ms, on an Internet-scale graph with billions of edges. This is far beyond the current state of the art, due to the immense computational cost of processing a PPR query. We achieve this goal with a novel algorithm kPAR, which utilizes the massive parallel processing power of GPUs. The main challenge in designing a GPU-based PPR algorithm lies in that a GPU is mainly a parallel computation device, whereas PPR processing involves graph traversals and value propagation operations, which are inherently sequential and memory-bound. Existing scalable PPR algorithms are mostly described as single-thread CPU solutions that are resistant to parallelization. Further, they usually involve complex data structures which do not have efficient adaptations on GPUs. kPAR overcomes these problems via both novel algorithmic designs (namely, adaptive forward push and inverted random walks ) and system engineering (e.g., load balancing) to realize the potential of GPUs. Meanwhile, kPAR provides rigorous guarantees on both result quality and worst-case efficiency. Extensive experiments show that kPAR is usually 10x faster than parallel adaptations of existing methods. Notably, on a billion-edge Twitter graph, kPAR answers a top-1000 PPR query in 42.4 milliseconds.

Cloud service recommendation based on unstructured textual information

With the rapid growth of cloud service platform, it is more and more difficult for users to discover cloud services satisfying their personalized demands. The unstructured textual information, such as the service descriptive texts and service tags, contain rich features of cloud services and are useful for cloud service platforms to construct personalized service recommendation. This paper proposes a two-stage model for cloud service recommendation by integrating the information of service descriptive texts and service tags. In the first stage, we propose a Hierarchical Dirichlet Processes (HDP) model to cluster cloud services into an optimal number of clusters based on descriptive texts. In the second stage, we propose a Personalized PageRank algorithm based on service tags to rank and recommend cloud services in each cluster. Our experiments on a real data set show that the proposed two-stage model can segment cloud services well and obtain accurate recommendation results compared with the baseline methods.

Cognitive color palette creation using client message and color psychology

Color psychology is the study of the effect of colors on human behavior. This area of study is interesting and challenging because there is no simple “one-size-fits all” mapping between a color and the message that color can evoke in a particular person. Nevertheless, many of our daily decisions are influenced by the color palettes being presented to us. In this paper, we propose a novel computational algorithm to create color palettes that convey certain messages. It starts by constructing a weighted graph of color categories and messages that are based on semantic similarities between vector representation of the messages and delta-E color similarities between colors. The color selection process for a given message includes applying a Personalized PageRank algorithm on the graph, treating the message nodes as the personalizing factors. As a result, we have a distribution of probabilities over all the color nodes that have higher probabilities for the nodes neighboring to given messages. Finally, palettes are ranked on the basis of visual aesthetics, novelty, and the conflicting/reinforcing messages they evoke. We have applied this work to several interesting use cases where artists and fashion designers used the resulting color palettes to assist their creation process.

A Survey on Personalized PageRank Computation Algorithms

Personalized PageRank (PPR) is an important variation of PageRank, which is a widely applied popularity measure for Web search. Unlike the original PageRank, PPR is a node proximity measure that represents the degree of closeness among multiple nodes within a graph. It is also widely applied to diverse domains, such as information retrieval, recommendations, and knowledge discovery, due to its theoretical simplicity and flexibility. However, computing PPR in large graphs using naive algorithms such as iterative matrix multiplication and matrix inversion is not fast enough for many of these applications. Therefore, devising efficient PPR algorithms has been one of the most important subjects in large-scale graph processing. In this paper, we review the algorithms for efficient PPR computations, organizing them into five categories based on their core ideas. Along with detailed explanations including recent advances and their applications, we provide a multifaceted comparison of the algorithms.

An Efficient and Adaptive Method for Overlapping Community Detection in Real‐World Networks

In real-world networks, nodes may belong to more than one community simultaneously. Overlapping community detection in complex networks is a challenging task. An adaptive overlapping community detection method based on seed selection and expansion is proposed. Depending on the restrictions on the seed selection stage, a set of seeds is generated without specified set size. The personalized PageRank algorithm is used to evaluate the community for seed expansion. The uncovered nodes could be adaptively allocated to the appropriate clusters. A thorough comparison between the proposed method and other overlapping community detection methods considered is provided to indicate the effectiveness of the former. The experimental results demonstrate that the presented method is effective.

Co-occurrence graphs for word sense disambiguation in the biomedical domain.

Word sense disambiguation is a key step for many natural language processing tasks (e.g. summarization, text classification, relation extraction) and presents a challenge to any system that aims to process documents from the biomedical domain. In this paper, we present a new graph-based unsupervised technique to address this problem. The knowledge base used in this work is a graph built with co-occurrence information from medical concepts found in scientific abstracts, and hence adapted to the specific domain. Unlike other unsupervised approaches based on static graphs such as UMLS, in this work the knowledge base takes the context of the ambiguous terms into account. Abstracts downloaded from PubMed are used for building the graph and disambiguation is performed using the personalized PageRank algorithm. Evaluation is carried out over two test datasets widely explored in the literature. Different parameters of the system are also evaluated to test robustness and scalability. Results show that the system is able to outperform state-of-the-art knowledge-based systems, obtaining more than 10% of accuracy improvement in some cases, while only requiring minimal external resources.