Coherent Dialog Generation with Query Graph

Learning to generate coherent and informative dialogs is an enduring challenge for open-domain conversation generation. Previous work leverage knowledge graph or documents to facilitate informative dialog generation, with little attention on dialog coherence. In this article, to enhance multi-turn open-domain dialog coherence, we propose to leverage a new knowledge source, web search session data, to facilitate hierarchical knowledge sequence planning, which determines a sketch of a multi-turn dialog. Specifically, we formulate knowledge sequence planning or dialog policy learning as a graph grounded Reinforcement Learning (RL) problem. To this end, we first build a two-level query graph with queries as utterance-level vertices and their topics (entities in queries) as topic-level vertices. We then present a two-level dialog policy model that plans a high-level topic sequence and a low-level query sequence over the query graph to guide a knowledge aware response generator. In particular, to foster forward-looking knowledge planning decisions for better dialog coherence, we devise a heterogeneous graph neural network to incorporate neighbouring vertex information, or possible future RL action information, into each vertex (as an RL action) representation. Experiment results on two benchmark dialog datasets demonstrate that our framework can outperform strong baselines in terms of dialog coherence, informativeness, and engagingness.

Similar Supergraph Search Based on Graph Edit Distance

Subgraph and supergraph search methods are promising techniques for the development of new drugs. For example, the chemical structure of favipiravir—an antiviral treatment for influenza—resembles the structure of some components of RNA. Represented as graphs, such compounds are similar to a subgraph of favipiravir. However, the existing supergraph search methods can only discover compounds that match exactly. We propose a novel problem, called similar supergraph search, and design an efficient algorithm to solve it. The problem is to identify all graphs in a database that are similar to any subgraph of a query graph, where similarity is defined as edit distance. Our algorithm represents the set of candidate subgraphs by a code tree, which it uses to efficiently compute edit distance. With a distance threshold of zero, our algorithm is equivalent to an existing efficient algorithm for exact supergraph search. Our experiments show that the computation time increased exponentially as the distance threshold increased, but increased sublinearly with the number of graphs in the database.

Supergraph Topology Feature Index for Personalized Interesting Subgraph Query in Large Labeled Graphs

Interesting subgraph query aims to find subgraphs that are isomorphic to the given query graph from a data graph and rank the subgraphs according to their interestingness scores. However, the existing subgraph query approaches are inefficient when dealing with large-scale labeled data graph. This is caused by the following problems: (i) the existing work mainly focuses on unweighted query graphs, while ignoring the impact of query constraints on query results. (ii) Excessive number of subgraph candidates or complex joins between nodes in the subgraph candidates reduce the query efficiency. To solve these problems, this paper proposes an intelligent solution. Firstly, an Isotype Structure Graph Compression (ISGC) strategy is proposed to compress similar nodes in a graph to reduce the size of the graph and avoid unnecessary matching. Then, an auxiliary data structure Supergraph Topology Feature Index (STFIndex) is designed to replace the storage of the original data graph and improve the efficiency of an online query. After that, a partition method based on Edge Label Step Value (ELSV) is proposed to partition the index logically. In addition, a novel Top-K interest subgraph query approach is proposed, which consists of the multidimensional filtering (MDF) strategy, upper bound value (UBV) (Size-c) matching, and the optimizational join (QJ) method to filter out as many false subgraph candidates as possible to achieve fast joins. We conduct experiments on real and synthetic datasets. Experimental results show that the average performance of our approach is 1.35 higher than that of the state-of-the-art approaches when the query graph is unweighted, and the average performance of our approach is 2.88 higher than that of the state-of-the-art approaches when the query graph is weighted.

Symmetric continuous subgraph matching with bidirectional dynamic programming

In many real datasets such as social media streams and cyber data sources, graphs change over time through a graph update stream of edge insertions and deletions. Detecting critical patterns in such dynamic graphs plays an important role in various application domains such as fraud detection, cyber security, and recommendation systems for social networks. Given a dynamic data graph and a query graph, the continuous subgraph matching problem is to find all positive matches for each edge insertion and all negative matches for each edge deletion. The state-of-the-art algorithm TurboFlux uses a spanning tree of a query graph for filtering. However, using the spanning tree may have a low pruning power because it does not take into account all edges of the query graph. In this paper, we present a symmetric and much faster algorithm SymBi which maintains an auxiliary data structure based on a directed acyclic graph instead of a spanning tree, which maintains the intermediate results of bidirectional dynamic programming between the query graph and the dynamic graph. Extensive experiments with real and synthetic datasets show that SymBi outperforms the state-of-the-art algorithm by up to three orders of magnitude in terms of the elapsed time.

Subgraph-Indexed Sequential Subdivision for Continuous Subgraph Matching on Dynamic Knowledge Graph

Continuous subgraph matching problem on dynamic graph has become a popular research topic in the field of graph analysis, which has a wide range of applications including information retrieval and community detection. Specifically, given a query graph q , an initial graph G 0 , and a graph update stream △ G i , the problem of continuous subgraph matching is to sequentially conduct all possible isomorphic subgraphs covering △ G i of q on G i (= G 0 ⊕ △ G i ). Since knowledge graph is a directed labeled multigraph having multiple edges between a pair of vertices, it brings new challenges for the problem focusing on dynamic knowledge graph. One challenge is that the multigraph characteristic of knowledge graph intensifies the complexity of candidate calculation, which is the combination of complex topological and attributed structures. Another challenge is that the isomorphic subgraphs covering a given region are conducted on a huge search space of seed candidates, which causes a lot of time consumption for searching the unpromising candidates. To address these challenges, a method of subgraph-indexed sequential subdivision is proposed to accelerating the continuous subgraph matching on dynamic knowledge graph. Firstly, a flow graph index is proposed to arrange the search space of seed candidates in topological knowledge graph and an adjacent index is designed to accelerate the identification of candidate activation states in attributed knowledge graph. Secondly, the sequential subdivision of flow graph index and the transition state model are employed to incrementally conduct subgraph matching and maintain the regional influence of changed candidates, respectively. Finally, extensive empirical studies on real and synthetic graphs demonstrate that our techniques outperform the state-of-the-art algorithms.

Graph Pattern Matching: Capturing Bisimilar Subgraph

Graph pattern matching that aims to seek out answer graphs in a data graph matching a provided graph, plays a fundamental role as a part of graph search for graph databases. “Matching” indicates that the two graphs are correlated, such as bisimulation, isomorphism, simulation, etc. The strictness of bisimulation is between simulation and isomorphism. Seldom work has been done to search for bisimulation subgraphs. This research focuses on the problem. The symbol [Formula: see text] is introduced to fundamental modal logic language, thereby yielding [Formula: see text] language; the symbols [Formula: see text] is added for forming [Formula: see text] formulas. Then conclusions about graph bisimulations are shown. Subsequently, a theorem with detailed proof is presented, stating that [Formula: see text] formulas characterize finite directed graphs modulo bisimulation. According to the conclusions and theorem, algorithms for finding subgraphs are proposed. After dividing the query graph, the match graphs undergo the characterization using [Formula: see text] formulas. In the data graphs, by model checking the formulas, the answer graphs exhibiting bisimilarity to the match graphs are able to be captured.