Effective Tag Recommendation System Based on Topic Ontology UsingWikipediaand WordNet

Problem statement: Ontological user profile generation was a semantic approach to derive richer concept based user profiles. It depends on the semantic relationship of concepts. This study focuses on ontology to derive concept oriented user profile based on user search queries and clicked documents.This study proposes concept based on topic ontology which derives the concept based user profiles more independently. It was possible to improve the search engine processes more efficiently. Approach: This process consists of individual user’s interests, topical categories of user interests and identifies the relationship among the concepts. The proposed approach was based on topic ontology for concept based user profile generation from search engine logs. Spreading activation algorithm was used to optimize the relevance of search engine results. Topic ontology was constructed to identify the user interest by assigning activation values and explore the topics similarity of user preferences. Results: To update and maintain the interest scores, spreading activation algorithm was proposed. User interest may change over the period of time which was reflected to user profiles. According to profile changes, search engine was personalized by assigning interest scores and weight to the topics. Conclusion: Experiments illustrate the efficacy of proposed approach and with the help of topic ontology user preferences can be identified correctly. It improves the quality of the search engine personalization by identifying the user’s precise needs.

Addressed in this paper is the issue of semantic relationship extraction from semi-structured documents. Many research efforts have been made so far on the semantic information extraction. However, much of the previous work focuses on detecting `isolated' semantic information by making use of linguistic analysis or linkage information in web pages and limited research has been done on extracting semantic relationship from the semi-structured documents. In this paper, we propose a method for semantic relationship extraction by using the logical information in the semi-structured document (semi-structured document usually has various types of structure information, e.g. a semi-structured document may be hierarchical laid out). To the best of our knowledge, extracting semantic relationships by using logical information has not been investigated previously. A probabilistic approach has been proposed in the paper. Features used in the probabilistic model have been defined.

Tag recommendation approaches have historically focused on maximizing the relevance of the recommended tags for a given object, such as a movie or a song. Nevertheless, different users may be interested in the same object for different reasons---for instance, the Star Wars movies may appeal to both adventure as well as to fantasy movie fans. In this situation, a sensible strategy is to provide a user with diverse recommendations of how to tag the object. In this paper, we address the problem of recommending relevant and diverse tags as a ranking problem. In particular, we propose a novel tag recommendation approach that explicitly takes into account the possible topics (e.g., categories) underlying an object in order to promote tags with high coverage and low redundancy with respect to these topics. We thoroughly evaluate our proposed approach using data collected from two popular Web 2.0 applications, namely, LastFM and MovieLens. Our experimental results attest the effectiveness of our approach at promoting more relevant and diverse tags in contrast to state-of-the-art relevance-based methods as well as a recently proposed method that takes both relevance and diversity into account.

Past studies have proposed solutions that analyze Stack Overflow content to help users find desired information or aid various downstream software engineering tasks. A common step performed by those solutions is to extract suitable representations of posts; typically, in the form of meaningful vectors. These vectors are then used for different tasks, for example, tag recommendation, relatedness prediction, post classification, and API recommendation. Intuitively, the quality of the vector representations of posts determines the effectiveness of the solutions in performing the respective tasks. In this work, to aid existing studies that analyze Stack Overflow posts, we propose a specialized deep learning architecture Post2Vec which extracts distributed representations of Stack Overflow posts. Post2Vec is aware of different types of content present in Stack Overflow posts, i.e., title, description, and code snippets, and integrates them seamlessly to learn post representations. Tags provided by Stack Overflow users that serve as a common vocabulary that captures the semantics of posts are used to guide Post2Vec in its task. To evaluate the quality of Post2Vec's deep learning architecture, we first investigate its end-to-end effectiveness in tag recommendation task. The results are compared to those of state-of-the-art tag recommendation approaches that also employ deep neural networks. We observe that Post2Vec achieves 15-25 percent improvement in terms of F1-score@5 at a lower computational cost. Moreover, to evaluate the value of representations learned by Post2Vec, we use them for three other tasks, i.e., relatedness prediction, post classification, and API recommendation. We demonstrate that the representations can be used to boost the effectiveness of state-of-the-art solutions for the three tasks by substantial margins (by 10, 7, and 10 percent in terms of F1-score, F1-score, and correctness, respectively). We release our replication package at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/maxxbw/Post2Vec</uri> .

Tag recommendation is an integral part of any bookmarking application. With the growing popularity in Web 2.0 usage, recommending tags is of utmost importance in enabling a user to perform bookmarking easily. An issue that most recommendation systems do not consider is that users have a tendency to choose from tags that are suggested to them, which might bias the true popular rankings of tags. In this paper we consider the problem of tag recommendation for bookmarks based on user feedback. We propose an approach for automatic tag recommendation by using a suggestion set and discuss a randomized suggestion rule for learning the true popular tags. As our algorithm depends on the frequency of tag suggestions, the action of spammers and malicious users may result in skewed ranking of tags for a bookmark. Hence, there arises a need to identify malicious users and spam posts to reinforce the efficiency of our algorithm because spammers can easily mislead a system. We have proposed an approach for classifying spammers based on tagging history of users. Our approach basically estimates the probability of a user being a spammer by analyzing previous posts and tags. Our analysis on a dataset of a popular bookmarking site shows that the proposed method is effective in suggesting the true popular tags and identifying spammers.

One of the primary goals of tag recommendation approaches is to deal with the problem of ambiguity of tags in a folksonomy by helping users to select the most appropriate tag to annotate a resource. We propose in this work, an original approach for tag recommendation applying a link prediction using supervised machine learning. Given a user (target user) and a resource (target resource) the proposed algorithm computes a list of tags best suited for recommending target user to annotate the target resource. It first searches for users similar to the target user. Then a link prediction approach is applied on a temporal sequence of bipartite graphs coding the history of tagging of retrieved similar users. This results in obtaining one or more lists of tags for the target resource or similar resources. These lists are then merged using a list aggregation method to get a single list of tags for recommendation. The first prototype of this approach is described in this article. Preliminary results of applying the proposed approach to real dataset extracted from the bibliographical folksonomy CiteULike1 show the validity of the approach.

This paper presents an approach to construct rules automatically to identify semantic relations, such as Hypernym, Part_Of and Material from Modern Chinese Standard Dictionary (MCSD). This approach combines all the useful information such as part of speeches, syntactic information, and extracted semantic relations together in order to identify as many semantic relations as possible. However, not all the information provided, like syntactic information, is reliable. A method is required to ensure that the right information up to a point can be picked. And then all possible semantic relations are constructed as a net after being identified. Finally, we evaluate the net by determining word similarity.

Geographic information integration and fusion from multi-sources geospatial data is still a challenge because of semantic heterogeneity of geospatial entity data. These will delay geospatial cooperation decision support and integrated assessment application for natural resources and environmental problem. Ontology as a conceptualization and knowledge representation of application domain have provided a potential and aided support for entity matching and information integration of multiple sources and heterogeneous data. However, many data sources may not provided ontology definition and formalization representation. This paper will provide a method of graph model based ontologies representation and ontologies learning from data sources schema. Entity matching and fusion with ontology support is also discussed. This paper has classified unary properties and binary properties in ontology representation. Binary semantic relations in ontologies are analyzed. Based on semantic relations, the Tree model and Network model of ontology representation are defined. Through analyzing entity data schema, semantic relations extraction and learning are also discussed in constitution of entity ontologies. The framework of entity fusion with ontology support has been designed. Finally entity matching and fusion methods are discussed.

A large volume of unstructured data, especially text data, is generated and exchanged daily. Consequently, the importance of extracting patterns and discovering knowledge from textual data is significantly increasing. As the task of automatically recognizing the relations between two or more entities, semantic relation extraction has a prominent role in the exploitation of raw text. This article surveys different approaches and types of relation extraction in English and the most prominent proposed methods in Persian. We also introduce, analyze, and compare the most important datasets available for relation extraction in Persian and English. Furthermore, traditional and emerging evaluation metrics for supervised, semi-supervised, and unsupervised methods are described, along with pointers to commonly used performance evaluation datasets. Finally, we briefly describe challenges in extracting relationships in Persian and English and dataset creation challenges.

Hierarchical learning strategy in semantic relation extraction

Semantic relation extraction has been a known branch in Natural language processing allowing the identification of relations between words. Knowledge-based approaches including wordnet and pattern matching approaches such as Hearst for hypernyms are used among many others. The search for relations is limited to sentences which avert the rest of the text narrowing the semantic relation extraction. In this paper, we introduce the use of context for improving the extraction using global information from the text and domain ontology.

The ontolex interface is anchored on two essential elements of human knowledge: conceptual structure and lexical access. A well-structured ontolex interface facilitates bilingual representation of both the conceptual structure and the lexical information which will be crucial in overcoming linguistic barriers. In this chapter we propose a robust and versatile approach for constructing the ontolex interface infrastructure: Bilingual Ontological WordNet (BOW). The Academia Sinica Bilingual Ontological WordNet (Sinica BOW) was constructed in 2003 using this approach. We argue that this novel combination of ontology and WordNet will (1) give each linguistic form a precise location in the taxonomy, (2) clarify the relation between the conceptual classification and its linguistic instantiation, and (3) facilitate a genuine crosslingual access of knowledge.

Extracting semantic relations from the Quranic Arabic based on Arabic conjunctive patterns

Ontologies play an important role of the applications in recent years, especially on the Semantic Web, Information Retrieval, Information Extraction, and Question Answering. Ontologies are used in order to specify the knowledge that is exchanged and shared between the different systems. Ontologies define the formal semantics of the terms used for describing data, and the relations between these terms. Hence, one of the important steps for building the domain specific ontology is to construct the semantic relations among the terms of the ontology. This paper represents our proposed method for automatic semantic relation extraction from text documents of the ACM Digital Library. A random sample among 170 categories of ACM categories is used to evaluate our proposed method. Results generated show that our proposed method achieves high precision.

Purpose Considering the shortcomings of the ASIS&amp;T Web-based thesaurus in representing concepts and semantic relations, there is a need to use more effective semantic tools, such as ontologies. The purpose of this paper is to build a prototype ontology (ASIS&amp;TOnto) based on a Web-based thesaurus. Design/methodology/approach The prototype ontology was built based on the ASIS&amp;T Web-based thesaurus by using the methodology developed by Noy and McGuinness (2001) and with the use of Protégé 4.3.3. With regard to the purpose of the study and the massive amount of concepts represented in the ASIS&amp;T thesaurus, the focus of term selection for creating the core source of vocabulary for the prototype ontology was the “searching” area. The knowledge engineering approach (Na and Neoh, 2008) was used to extract concepts and semantic relations. The criterion for extracting semantic relations from each concept pair was one-sentence statement, one-paragraph statement and one-page statement. Finally, the extracted relations were analysed by subject experts. Findings Based on the findings of this research, the possibility of using the methodology developed by Noy and McGuinness (2001) for building an ontology based on a Web-based thesaurus in the field of LIS was investigated and the prototype ontology (ASIS&amp;TOnto) was constructed in the area of “searching”. Practical implications The primary implication of ASIS&amp;TOnto is aligned with Web 3.0 research where implications of semantic modelling are a priority for community equity of access to information as a basic human right. ASIS&amp;TOnto ends up with an ontology that is comprehensive, at least with respect to expressivity of the current ASIS&amp;T thesaurus user interface definition languages; that is, it can be universal and can be extendable to future user interfaces that do not exist at the moment. Originality/value For the first time, a new enriched semantic tool (ASIS&amp;TOnto) was constructed based on the ASIS&amp;T Web-based thesaurus.