Semantic Matching Model Research Articles

Aligned variational autoencoder for matching danmaku and video storylines

We study a task of aligning time-sync video comments (danmaku) to narrative video storylines, which is helpful for finding semantic segmentation of videos and conducting fine-grained user feedback analyses. Due to the mismatch of text styles and the shift of topics, it is hard to apply previous semantic matching models directly for the alignment. To tackle the problem, we propose to utilize variational auto-encoders to map both user comments and storylines into latent spaces. By posing a matching loss on their latent codes, we reduce their mismatches in the latent space and make the alignment easier to learn. To handle constraints in the alignment, we also apply dynamic programming for finding global optimal outputs. According to experiments on a TV series dataset (consisting of about 10 K pairs of storylines and danmaku sent by users), the proposed model can achieve competitive performances.

A deep embedding model for knowledge graph completion based on attention mechanism

Knowledge graph completion has become a well-studied problem and a non-trivial task with the broad application of the knowledge graphs. Previously, a lot of works have been proposed to solve the knowledge graph completion problem, for example, a series of Trans model, semantic matching models, convolutional neural networks based methods and so on. However, a series of Trans models and semantic matching models only focused on the shadow information of the knowledge graph, thus failed to capture the implicit fine-grained feature in the triple of knowledge graphs; convolutional neural networks based methods learned more expressive feature for knowledge graph completion, and it also ignored the directional relation characteristic and implicit fine-grained feature in the triple. In this paper, we propose a novel knowledge graph completion model named directional multi-dimensional attention convolution model that explores directional information and an inherent deep expressive characteristic of the triple. At last, we evaluate our directional multi-dimensional attention convolution model based on three standard evaluation criteria in two robust datasets, and the experiment shows that our model achieves state-of-the-art MeanRank.

An ontology-based deep learning approach for triple classification with out-of-knowledge-base entities

Knowledge graphs (KGs) are one of the most common frameworks for knowledge representation. However, they suffer from a severe scalability problem that hinders their usage. KG embedding aims to provide a solution to this issue. Nonetheless, general approaches are incapable of representing and reasoning about information not previously contained in the graph. This paper proposes to leverage semantic and ontological information for a significant benefit of knowledge graph completion, focusing on triple classification. The goal of this task is to determine whether a given fact holds. Furthermore, this paper also considers the classification of facts that include entities that have not been seen during training, denoted out-of-knowledge-base or OOKB entities. An incremental method is presented, composed of six stages. Although the proposal can be applied to any KG embedding model, this work focuses on its application for semantic matching models, such as ComplEx and DistMult. Compared to other approaches, our proposal is model-agnostic, computationally inexpensive, and does not require retraining. The results show that triple classification accuracy scales up to 15% with the proposed approach, as well as accelerating the convergence of the model to its optimal solution. Furthermore, facts containing OOKB entities can be classified with a reasonable accuracy.

DAM: Transformer-based relation detection for Question Answering over Knowledge Base

Relation Detection is a core component of Knowledge Base Question Answering (KBQA). In this paper, we propose a Transformer-based deep attentive semantic matching model (DAM), to identify the KB relations corresponding to the questions. The DAM is completely based on the attention mechanism and applies the fine-grained word-level attention to enhance the matching of questions and relations. On the basis of the DAM, we build a three-stage KBQA pipeline system. The experimental results on multiple benchmarks demonstrate that our DAM model outperforms previous methods on relation detection. In addition, our DAM-based KBQA system also achieves state-of-the-art results on multiple datasets.

An intelligent personalized web blog searching technique using fuzzy-based feedback recurrent neural network

In web information retrieval, every single user has a unique contextual objective when searching for information on web blogs. Based on the given query, the task of the web search engine is to fetch the most related information from the collection of web blogs. In order to enhance the searching ability, efficient semantic matching models, richer training, and evaluation resources are required. The conventional keyword-based search algorithms have a minimum efficiency in knowing users’ intentions compared to machine learning algorithms. Recently, neural networks are well-recognized in information retrieval due to the ability of vector representation learning. This paper proposes an adaptive fuzzy feedback recurrent neural network-based web blog searching technique which follows inverse filtering (IF) algorithm using Word2Vec representation. Initially, the user query is pre-processed, and then given to the IF for accelerating the search process. Inside the IF, the web blog content is labelled according to their blog information and stored in the hash table, and then relevant contents are extracted by TF–IDF; this result is given to the similarity estimation to obtain the similarity score. Finally, the proposed technique considers the feedback got from users and performs re-ranking by fuzzy-based RNN to achieve rich user intention satisfaction. The implementation is performed in Python, and statistical measures such as accuracy, precision, and recall are utilized for performance evaluation. The results demonstrate that the proposed technique has enhanced accuracy (94%) compared to conventional techniques, namely deep auto-encoder, deep neural networks, and artificial neural networks.

Improve the translational distance models for knowledge graph embedding

Knowledge graph embedding techniques can be roughly divided into two mainstream, translational distance models and semantic matching models. Though intuitive, translational distance models fail to deal with the circle structure and hierarchical structure in knowledge graphs. In this paper, we propose a general learning framework named TransX-pa, which takes various models (TransE, TransR, TransH and TransD) into consideration. From this unified viewpoint, we analyse the learning bottlenecks are: (i) the common assumption that the inverse of a relation r is modelled as its opposite − r; and (ii) the failure to capture the rich interactions between entities and relations. Correspondingly, we introduce position-aware embeddings and self-attention blocks, and show that they can be adapted to various translational distance models. Experiments are conducted on different datasets extracted from real-world knowledge graphs Freebase and WordNet in the tasks of both triplet classification and link prediction. The results show that our approach makes a great improvement, showing a better, or comparable, performance with state-of-the-art methods.

Forum Duplicate Question Detection by Domain Adaptive Semantic Matching

Community Question Answering (CQA) forums, such as Stack Overflow, Stack Exchange and Massive Open Online Course (MOOC) forums, spend a lot of manpower and time to manage duplicate questions on the forum. Mismatch of duplicate questions makes users keep asking “new” questions, and the continuous accumulation of duplicate questions may interfere with their information searching again, affecting user satisfaction. Neural Networks (NN) models for parsing semantics provide the possibility of end-to-end duplicate question detection. Whereas, due to lack of domain data and expertise, NN models for semantic parsing are rarely directly applied to CQA duplicate question detection. This paper proposes a Semantic Matching Model (SMM) integrated with the multi-task transfer learning framework for multi-domain forum duplicate question detection. By designing the word-to-sentence interaction mechanism based on the word-to-word interaction, SMM can automatically choose to ignore or pay attention to potential similar words according to the semantics at the sentence level. The experiments on the benchmark data set and MOOC forum data set state that SMM outperforms baselines, its interaction mechanism is effective and it has an advantage in cross-domain duplicate question detection.

Knowledge Graph Completion: A Review

Knowledge graph completion (KGC) is a hot topic in knowledge graph construction and related applications, which aims to complete the structure of knowledge graph by predicting the missing entities or relationships in knowledge graph and mining unknown facts. Starting from the definition and types of KGC, existing technologies for KGC are analyzed in categories. From the evolving point of view, the KGC technologies could be divided into traditional and representation learning based methods. The former mainly includes rule-based reasoning method, probability graph model, such as Markov logic network, and graph computation based method. The latter further includes translation model based, semantic matching model based, representation learning based and other neural network model based methods. In this article, different KGC technologies are introduced, including their advantages, disadvantages and applicable fields. Finally the main challenges and problems faced by the KGC are discussed, as well as the potential research directions.

An image retrieval method based on semantic matching with multiple positional representations

Text-based image retrieval requires manual annotation or automatic labeling of the machine. Manual annotation is time-consuming, and simple text description is difficult to fully express the content of the image. Existing deep models rely on the representation of a single sentence, and such methods cannot well capture the contextualized local information in the matching process. In response to these problems, this paper presents a new retrieval idea based on image caption. First, the image description sentences of images are generated by using the image caption model. Then, for the sentence matching model, we propose a multiple positional representations semantic matching model. We use two interrelated Bi-LSTMs and the attention mechanism to match sentences. the matching score is finally produced by aggregating interactions between these different positional sentence representations. The sentence matching model is used to match the retrieval sentence with the image description sentences in the image library. In our experiments, the accuracy of the proposed image caption model and the sentence matching model are all improved compared with the competitive models, and our method can complete the image retrieval task.

Fine-Resolution Population Mapping from International Space Station Nighttime Photography and Multisource Social Sensing Data Based on Similarity Matching

Previous studies have attempted to disaggregate census data into fine resolution with multisource remote sensing data considering the importance of fine-resolution population distribution in urban planning, environmental protection, resource allocation, and social economy. However, the lack of direct human activity information invariably restricts the accuracy of population mapping and reduces the credibility of the mapping process even when external facility distribution information is adopted. To address these problems, the present study proposed a novel population mapping method by combining International Space Station (ISS) photography nighttime light data, point of interest (POI) data, and location-based social media data. A similarity matching model, consisting of semantic and distance matching models, was established to integrate POI and social media data. Effective information was extracted from the integrated data through principal component analysis and then used along with road density information to train the random forest (RF) model. A comparison with WordPop data proved that our method can generate fine-resolution population distribution with higher accuracy ( R 2 = 0.91 ) than those of previous studies ( R 2 = 0.55 ). To illustrate the advantages of our method, we highlighted the limitations of previous methods that ignore social media data in handling residential regions with similar light intensity. We also discussed the performance of our method in adopting social media data, considering their characteristics, with different volumes and acquisition times. Results showed that social media data acquired between 19:00 and 8:00 with a volume of approximately 300,000 will help our method realize high accuracy with low computation burden. This study showed the great potential of combining social sensing data for disaggregating fine-resolution population.

Combining Fact Extraction and Verification with Neural Semantic Matching Networks

The increasing concern with misinformation has stimulated research efforts on automatic fact checking. The recentlyreleased FEVER dataset introduced a benchmark factverification task in which a system is asked to verify a claim using evidential sentences from Wikipedia documents. In this paper, we present a connected system consisting of three homogeneous neural semantic matching models that conduct document retrieval, sentence selection, and claim verification jointly for fact extraction and verification. For evidence retrieval (document retrieval and sentence selection), unlike traditional vector space IR models in which queries and sources are matched in some pre-designed term vector space, we develop neural models to perform deep semantic matching from raw textual input, assuming no intermediate term representation and no access to structured external knowledge bases. We also show that Pageview frequency can also help improve the performance of evidence retrieval results, that later can be matched by using our neural semantic matching network. For claim verification, unlike previous approaches that simply feed upstream retrieved evidence and the claim to a natural language inference (NLI) model, we further enhance the NLI model by providing it with internal semantic relatedness scores (hence integrating it with the evidence retrieval modules) and ontological WordNet features. Experiments on the FEVER dataset indicate that (1) our neural semantic matching method outperforms popular TF-IDF and encoder models, by significant margins on all evidence retrieval metrics, (2) the additional relatedness score and WordNet features improve the NLI model via better semantic awareness, and (3) by formalizing all three subtasks as a similar semantic matching problem and improving on all three stages, the complete model is able to achieve the state-of-the-art results on the FEVER test set (two times greater than baseline results).1

CAD assembly model retrieval based on multi-source semantics information and weighted bipartite graph

Content-based CAD assembly model retrieval focusing on similarity measure of bottom-level geometry and topology information can hardly meet design demand of engineering designers in the product design reuse. To search quickly and effectively the CAD assembly models with specific functions by using requirements of mechanical product design, inspire designers’ design thought and product design innovation, an assembly retrieval approach based on multi-source semantics information and weighted bipartite graph is proposed. The approach concerns the high-level semantics information rather than geometry and topology information of assembly model. First, the semantic ontology for mechanical assembly model is constructed considering multi-source semantics information including part semantics, assembly constraint semantics and functional semantics. Function region extraction algorithm is given for identifying functional structures in assembly model, and functional semantics of assembly model is annotated from the aspects of assembly structure granularity and part shape granularity. Subsequently, semantic similarity between concepts is evaluated comprehensively with WordNet and domain ontology, and a semantic matching model is built based on weighted bipartite graph (WBG) for solving the many-to-many semantics matching problem of semantic retrieval for CAD assembly models. Finally, a semantic-based assembly retrieval prototype system is developed and the experiment cases verify effectiveness and feasibility of the proposed assembly retrieval approach.

Community-Based Question Answering via Asymmetric Multi-Faceted Ranking Network Learning

Nowadays the community-based question answering (CQA) sites become the popular Internet-based web service, which have accumulated millions of questions and their posted answers over time. Thus, question answering becomes an essential problem in CQA sites, which ranks the high-quality answers to the given question. Currently, most of the existing works study the problem of question answering based on the deep semantic matching model to rank the answers based on their semantic relevance, while ignoring the authority of answerers to the given question. In this paper, we consider the problem of community-based question answering from the viewpoint of asymmetric multi-faceted ranking network embedding. We propose a novel asymmetric multi-faceted ranking network learning framework for community-based question answering by jointly exploiting the deep semantic relevance between question-answer pairs and the answerers' authority to the given question. We then develop an asymmetric ranking network learning method with deep recurrent neural networks by integrating both answers' relative quality rank to the given question and the answerers' following relations in CQA sites. The extensive experiments on a large-scale dataset from a real world CQA site show that our method achieves better performance than other state-of-the-art solutions to the problem.

A probabilistic ranking framework for web-based relational data imputation

Due to richness of information on web, there is an increasing interest to search for missing attribute values in relational data on web. Web-based relational data imputation has to first extract multiple candidate values from web and then rank them by their matching probabilities. However, effective candidate ranking remains challenging because web documents are unstructured and popular search engines can only provide with relevant but not necessarily semantically matching information.In this paper, we propose a novel probabilistic approach for ranking the web-retrieved candidate values. It can integrate various influence factors, e.g. snippet rank order, occurrence frequency, occurrence pattern, and keyword proximity, in a single framework by semantic reasoning. The proposed framework consists of snippet influence model and semantic matching model. The snippet influence model measures the influence of a snippet, and the semantic matching model measures the semantic similarity between a candidate value in a snippet and a missing relational value in a tuple. We also present effective probabilistic estimation solutions for both models. Finally, we empirically evaluate the performance of the proposed framework on real datasets. Our extensive experiments demonstrate that it outperforms the state-of-the-art techniques by considerable margins on imputation accuracy.

A distributed semantic filtering model based on approximate automata for heterogeneous multi-sensor networks

An increasing number of sensor networks are deployed to collect information on a wide range of applications. However, these sensor networks are isolated to each other. Meanwhile, processing large-scale of data stream is the neck of the push-mode system. To match the huge event stream in heterogeneous sensor network data, we propose approximation semantic factor ASR, according to which we further propose the approximately semantic matching model for heterogeneous multi-sensor networks called approximately semantic matching model for heterogeneous multi-sensor networks ASMMN. Furthermore, the evaluate pattern algorithm is put forward. ASMMN can effectively solve the semantic coupling and speed matching. Simulation results show that the proposed model can accelerate the matching process and reduce memory consumption of intermediate states among event streams. Meanwhile, the model has a good scalability that can be widely used in the heterogeneous push-mode sensor networks and internet of things IOTs.

Pure mediated priming: A retrospective semantic matching model.

Mediated priming refers to the activation of a target (e.g., stripes) by a prime (e.g., lion) that is related indirectly via a connecting mediator (e.g., tiger). In previous mediated priming studies (e.g., McNamara & Altarriba, 1988), the mediator was associatively related to the prime. In contrast, pure mediated priming (e.g., spoon --> can) lacks a strong association between prime and mediator (e.g., spoon --> soup) and between mediator and target (e.g., soup --> can). This study establishes the existence of pure mediated priming and assesses which semantic priming model (spreading activation, compound-cue, or semantic matching) accounts for the results. Pure mediated priming occurred in 3 experiments across double and standard lexical decision tasks. However, such priming did not occur in a continuous lexical decision task, which precludes strategic processing. Overall, results indicate that a modified retrospective semantic matching model provides the best theoretical explanation of pure mediated priming. (PsycINFO Database Record (c) 2009 APA, all rights reserved).

Web service description and discovery based on semantic model

A novel semantic model of Web service description and discovery was proposed through an extension for profile model of Web ontology language for services (OWL-S) in this paper. Similarity matching of Web services was implemented through computing weighted summation of semantic similarity value based on specific domain ontology and dynamical satisfy extent evaluation for quality of service (QoS). Experiments show that the provided semantic matching model is efficient.

Inconsistency detection and resolution for context-aware middleware support

Context-awareness is a key feature of pervasive computing whose environments keep evolving. The support of context-awareness requires comprehensive management including detection and resolution of context inconsistency, which occurs naturally in pervasive computing. In this paper we present a framework for realizing dynamic context consistency management. The framework supports inconsistency detection based on a semantic matching and inconsistency triggering model, and inconsistency resolution with proactive actions to context sources. We further present an implementation based on the Cabot middleware. The feasibility of the framework and its performance are evaluated through a case study and a simulated experiment, respectively.