Figure 2 - uploaded by Phivos Mylonas
Content may be subject to copyright.
Source publication
Context modeling has long been acknowledged as a key aspect in a wide variety of problem domains. In this paper we focus on the combination of contextualization and personalization methods to improve the performance of personalized information retrieval. The key aspects in our proposed approach are (1) the explicit distinction between historic user...
Context in source publication
Context 1
... more concepts are considered, the context becomes narrower, that is, it contains less con- cepts and to smaller degrees. When the definition of context is extended to the case of fuzzy sets of concepts (Figure 2), the crisp taxonomic context C 0 is replaced by its fuzzy counterpart, that is, the fuzzy taxonomic context C. Obviously, the semantic meaning of fuzzy context remains the same as in the crisp case, that is, the above property must still hold. The context C of the nor- mal fuzzy set F on S is calculated as: ...
Similar publications
Short Messaging Service (SMS) is popu- larly used to provide information access to people on the move. This has resulted in the growth of SMS based Question An- swering (QA) services. However auto- matically handling SMS questions poses significant challenges due to the inherent noise in SMS questions. In this work we present an automatic FAQ-based...
While using semantic data can enable improved retrieval of suitable jobs or applicants in a recruitment process, cases of inconsistent or overly specific queries which would return no results still have to be dealt with. In this paper the extension of a semantic job portal with a novel query relaxation technique is presented which is able to return...
In this paper, an integrated information system is presented that offers enhanced search and retrieval capabilities to users of hetero-lingual digital audiovisual (a/v) archives. This innovative system exploits the advances in handlings a/v content and related metadata, as introduced by MPEG-4 and worked out by MPEG-7, to offer advanced services ch...
ABSTRACT The rapid increase of biomedical,literature available on the web,has made ,it increasingly ,difficult to find ,precise information. To implement an ,accurate ,biomedical information retrieval (IR) system, we must deal with the variants of biomedical terms carefully. In this paper, we focus on the generation of aliases, synonyms, acronyms,...
This paper presents a content-oriented video retrieval system which is capable of handling high volumes of content as well as various functionality requirements. It allows the audience to access the video contents based on their different interests in the selected video program. The retrieval system consists of a content-based scalable access platf...
Citations
... Retrieval in the wild can leverage many implicit and explicit context information [119][120][121][122][123][124][125][126][127][128][129][130][131][132]. The query "weather" may refer to the weather in Seattle or in London depending on where the user is located. ...
Neural networks with deep architectures have demonstrated significant performance improvements in computer vision, speech recognition, and natural language processing. The challenges in information retrieval (IR), however, are different from these other application areas. A common form of IR involves ranking of documents--or short passages--in response to keyword-based queries. Effective IR systems must deal with query-document vocabulary mismatch problem, by modeling relationships between different query and document terms and how they indicate relevance. Models should also consider lexical matches when the query contains rare terms--such as a person's name or a product model number--not seen during training, and to avoid retrieving semantically related but irrelevant results. In many real-life IR tasks, the retrieval involves extremely large collections--such as the document index of a commercial Web search engine--containing billions of documents. Efficient IR methods should take advantage of specialized IR data structures, such as inverted index, to efficiently retrieve from large collections. Given an information need, the IR system also mediates how much exposure an information artifact receives by deciding whether it should be displayed, and where it should be positioned, among other results. Exposure-aware IR systems may optimize for additional objectives, besides relevance, such as parity of exposure for retrieved items and content publishers. In this thesis, we present novel neural architectures and methods motivated by the specific needs and challenges of IR tasks.
... et. al, 2013;Dinh and Tamine 2012;Kebler et. al, 2009;Goker and Myrhaug, 2008;Vieira et. al, 2007). Device, user, task, document and spatio-temporal are the five context specific dimensions that have been explored in context-based information retrieval literature (Emmanouilidis et. al, 2013;Dinh and Tamine 2012;Li et. al, 2011;Asfari et. el, 2009;Mylonas et. al, 2008;Anand and Mobasher, 2007;Maeco et. al, 2013;Lukowic et. al, 2011;Zhou et. al, 2012). ...
When using Information Retrieval (IR) systems, users often present search queries made of ad-hoc keywords. It is then up to information retrieval systems (IRS) to obtain a precise representation of user's information need, and the context of the information. Context-aware ranking techniques have been constantly used over the past years to improve user interaction in their search activities for improved relevance of retrieved documents. Though, there have been major advances in context-adaptive systems, there is still a lack of technique that models and implements context-adaptive application. The paper addresses this problem using DROPT technique. The DROPT technique ranks individual user information needs according to relevance weights. Our proposed predictive document ranking model is computed as measures of individual user search in their domain of knowledge. The context of a query determines retrieved information relevance. Thus, relevant context aspects should be incorporated in a way that supports the knowledge domain representing users' interests. We demonstrate the ranking task using metric measures and ANOVA, and argue that it can help an IRS adapted to a user's interaction behaviour, using context to improve the IR effectiveness.
... -Context: user contexts are generally simulated by hypothetic context situations [104,128]. ...
... Context is simulated in References [13,128] by considering that an ODP 7 concept represents a potential user interest. In Reference [104] click-through data are also hypothesized and used as additional part of the simulated contexts; -Topics and tasks: topics could be prede ned [104], represent TREC topics as in Reference [28], or automatically generated using the top terms of the DMOZ (directory.mozilla) ontology [128] or the top terms of DMOZ ontology selected in a location branch of the DMOZ (US cities) in Reference [14]; DMOZ is an open-content directory of WWW links maintained and known as the Open Directory Project (ODP). ...
... Context is simulated in References [13,128] by considering that an ODP 7 concept represents a potential user interest. In Reference [104] click-through data are also hypothesized and used as additional part of the simulated contexts; -Topics and tasks: topics could be prede ned [104], represent TREC topics as in Reference [28], or automatically generated using the top terms of the DMOZ (directory.mozilla) ontology [128] or the top terms of DMOZ ontology selected in a location branch of the DMOZ (US cities) in Reference [14]; DMOZ is an open-content directory of WWW links maintained and known as the Open Directory Project (ODP). ...
Context such as the user’s search history, demographics, devices, and surroundings, has become prevalent in various domains of information seeking and retrieval such as mobile search, task-based search, and social search. While evaluation is central and has a long history in information retrieval, it faces the big challenge of designing an appropriate methodology that embeds the context into evaluation settings. In this article, we present a unified summary of a wide range of main and recent progress in contextual information retrieval evaluation that leverages diverse context dimensions and uses different principles, methodologies, and levels of measurements. More specifically, this survey article aims to fill two main gaps in the literature: First, it provides a critical summary and comparison of existing contextual information retrieval evaluation methodologies and metrics according to a simple stratification model; second, it points out the impact of context dynamicity and data privacy on the evaluation design. Finally, we recommend promising research directions for future investigations.
... From this work, we can see the first example of integration of Context Awareness and Document Ranking to improve IR. Later in 2007, on [36] the author proposed Ontology Knowledge Model to produce the Fuzzy Representation, then will be used in Clustering and Classification process for improving the performance of personalized IR [36]. The author has proposed a method for automatic extraction of Context Awareness factor such as persistent semantic user preferences, live and ad-hoc user interests, which are integrated with the Document Ranking in order to improve the accuracy and reliability of personalization and the ranking of the IR. ...
... From this work, we can see the first example of integration of Context Awareness and Document Ranking to improve IR. Later in 2007, on [36] the author proposed Ontology Knowledge Model to produce the Fuzzy Representation, then will be used in Clustering and Classification process for improving the performance of personalized IR [36]. The author has proposed a method for automatic extraction of Context Awareness factor such as persistent semantic user preferences, live and ad-hoc user interests, which are integrated with the Document Ranking in order to improve the accuracy and reliability of personalization and the ranking of the IR. ...
Most of the retrieved documents from the Information Retrieval (IR) System are irrelevant to the user because the IR cannot determine the user’s context. One of the main issues is that the relevancy of the retrieved documents is based on personal assessment that depends on the task to be done and its context. This paper provides the review of prior researches (2003–2016) and concludes the review by providing the summary of the research’s current trends, future direction and opportunity and defining the research gap. First, the findings show that in prior studies, there is no identification of contextual aspect has been done in optimizing the ranking function of the Malay IR. Second, in optimizing the ranking function, the integration process of context representation and document ranking must be done. This approach also has not been done yet in the development of Malay Document Retrieval. If it still stays in the current status, the Malay Document Retrieval system cannot be improved compared to the traditional languages of Context Aware IR System (English).
... Ph. Mylonas concerned with enhancing information retrieval by adding a level of personalization, where it needs to extract the users' preferences [15]. So, this framework proposed to use the user's query keywords to extract the implicit semantics and interests. ...
... The PWR approach realized high precision and improvement, and it is approximately close to Mylonas research [15] which achieved the highest precision and improvement average. The nonimprovement represented researches did not compare themselves to the search engine result like Waghmare [13], ZHOU [7], and Alonso [26]. ...
Continually, search engines improve their capabilities toward facilitating search and the retrieval enhancement. Despite the great efforts in the information retrieval field, the retrieved results may be out of user's expectation. This may be due to the huge number of web resources, and unidentified user's interests and domain. This paper proposes exploiting social annotations for improving retrieval based on personalization. The personalization focuses on web resources and retrieval process. In this context, new layer of knowledge is added to the web resource analysis and retrieval. Then, the additional knowledge leads to improve the retrieved results to be close to user's interests. So, it retrieved different results for the same query based on the user's interests. By applying the system, the experiments realize 36% precision improvement compared to non-personalized search engine. Moreover, the user satisfaction measured by evaluating search results versus user's priorities, where it was in between 92% up to 100%.
... Consequently, the term may be used under various different meanings. In a previous work [11] we have identified context as any information that might be used to specify the situation of an entity; the latter being a person, a place or an object that is relevant to the interaction between the user and the software system. In the medical framework this identification has to be adjusted accordingly, to tackle the nature of medical data that focus heavily on the temporal aspect of information (especially with respect to electronic medical records and biomedical information systems). ...
Scientific and technological knowledge and skills are becoming crucial for most data analysis activities. Two rather distinct, but at the same time collaborating, domains are the ones of computer science and medicine; the former offers significant aid towards a more efficient understanding of the latter’s research trends. Still, the process of meaningfully analyzing and understanding medical information and data is a tedious one, bound to several challenges. One of them is the efficient utilization of contextual information in the process leading to optimized, context-aware data analysis results. Nowadays, researchers are provided with tools and opportunities to analytically study medical data, but at the same time significant and rather complex computational challenges are yet to be tackled, among others due to the humanistic nature and increased rate of new content and information production imposed by related hardware and applications. So, the ultimate goal of this position paper is to provide interested parties an overview of major contextual information types to be identified within the medical data processing framework.
... For example, some returned results of the "pipe" concept from Google search engine describe the pipe as a hollow cylinder; some others describe it as a kind of music instrument and for others it is a set of data processing elements connected in series. Introducing the search context into the retrieval process offers a better understanding of the user's preferences and reduces obtrusiveness, inaccuracy, inconsistency, and distraction, by making it more contextrelevant and contextually coherent [26,34]. ...
Web Information retrieval aims to satisfy users' search needs which are highly dependent on their interests and preferences. In this context, web personalization provides an adapted information retrieval. Specific computational modeling is required to address the large amount of heterogeneous electronic documents and the lack of semantic annotations on the web which makes knowledge discovery challenging. In this paper, a novel system, based on fuzzy ontological user profile is proposed. The latter, is composed of history, positive and negative preferences. An implicit relevance judgments method is also introduced. Furthermore, the system is context-aware by integrating novel contextual similarity measures and supporting semantic fuzzification. Our proposal has been implemented and has endured a twofold evaluation. The results show that the proposed system can provide more personalized results and confirm the interest of the context-aware search.
... In the following, we first summarize the relevant applications about fuzzy ontologies, and then discuss other important issues of fuzzy ontologies. (Bobillo & Straccia, 2008) f-SHIF(D) √ √ Supporting that the degree of a fuzzy assertion is not only a constant, but also a variable DeLorean f-SROIQ(D) √ √ Reducing to crisp DLs to solve, and supporting fuzzy concrete domains D GURDL (Haarslev et al., 2007(Haarslev et al., , 2008 f-ALC √ Proposing some interesting techniques of optimization GERDS (Habiballa, 2007) f-ALC √ Adding role negation, top role, and bottom role to f-ALC FRESG f-ALC(G) √ √ Supporting fuzzy data information with customized fuzzy data types G YADLR (Stasinos & Georgios, 2007) SLG algorithm √ Allowing to deal with unknown degrees of truth in the fuzzy assertions of the knowledge base SoftFacts (Straccia, 2009a) SoftFacts An ontology-mediated top-k information retrieval system over relational databases LiFR (Tsatsou et al., 2014) f-DLP A lightweight fuzzy DL reasoner that supports a subset of fuzzy DL Programs (f-DLP) (Vallet et al., 2006;Zhou et al., 2006;Mylonas et al., 2008) The flexible nature of fuzzy ontology may support a wide range of approaches to the problems of retrieving relevant, appropriate, and most of all useful information which is a relevant key aspiration of research of the Semantic Web ...
... This allows ones to follow a unified approach to intelligent information retrieval, both for textual and multimedia documents. In addition, ontology-based context for personalized information retrieval was investigated in Vallet et al. (2006), Zhou et al. (2006), andMylonas et al. (2008). Mylonas et al. (2008) and Vallet et al. (2006) proposed methods for the automatic extraction of persistent semantic user preferences, and live, ad hoc user interests, which are combined in order to improve the accuracy and reliability of personalization for retrieval. ...
... In addition, ontology-based context for personalized information retrieval was investigated in Vallet et al. (2006), Zhou et al. (2006), andMylonas et al. (2008). Mylonas et al. (2008) and Vallet et al. (2006) proposed methods for the automatic extraction of persistent semantic user preferences, and live, ad hoc user interests, which are combined in order to improve the accuracy and reliability of personalization for retrieval. Zhou et al. (2006) showed a presentation and comprehensive retrieval framework, which incorporates a module to control the degree of personalization that is applied in the search result ranking, automatically adjusting it depending on the uncertainty contained in the search before personalization. ...
Ontology, as a standard (World Wide Web Consortium recommendation) for representing knowledge in the Semantic Web, has become a fundamental and critical component for developing applications in different real-world scenarios. However, it is widely pointed out that classical ontology model is not sufficient to deal with imprecise and vague knowledge strongly characterizing some real-world applications. Thus, a requirement of extending ontologies naturally arises in many practical applications of knowledge-based systems, in particular the Semantic Web. In order to provide the necessary means to handle such vague and imprecise information there are today many proposals for fuzzy extensions to ontologies, and until now the literature on fuzzy ontologies has been flourishing. To investigate fuzzy ontologies and more importantly serve as helping readers grasp the main ideas and results of fuzzy ontologies, and to highlight an ongoing research on fuzzy approaches for knowledge semantic representation based on ontologies, as well as their applications on various domains,
in this paper
,
we provide a comprehensive overview of fuzzy ontologies
. In detail, we
first
introduce fuzzy ontologies from the most common aspects such as
representation
(including categories, formal definitions, representation languages, and tools of fuzzy ontologies),
reasoning
(including reasoning techniques and reasoners), and
applications
(the most relevant applications about fuzzy ontologies). Then,
the other important issues
on fuzzy ontologies, such as
construction
,
mapping
,
integration
,
query
,
storage
,
evaluation
,
extension
, and
directions for future research
, are also discussed in detail. Also, we make some
comparisons and analyses
in our whole review.
... The result is a personalised domain ontology of the user's interest terms. The underlying algorithms used comprise probabilistic classifiers and Hopfield networks [28], fuzzy relational algebra [19] (the latter integrating expert knowledge by letting them attribute semantics to the discovered relations)[5]. ...
... Personal name profiling is a well studied problem in the context of Web mining, natural language processing, information extraction and personalized systems. There are wide varieties of applications to which personal name profiling can be helpful, such as personalized search [2], [3], information filtering [4], [5], recommender systems [5]- [8], semantic search engines [6], [9], [10], adaptive e-learning systems [11], intelligent tutoring systems [11]- [13], e-commerce applications [14], intelligent information retrieval [15], ads generation [16], active and passive help systems [17]- [19], knowledge management systems [20], [21], etc. ...
Personal information extraction as a subtask of entity profiling is the process of identifying and extracting useful and structured information of the focus person named-entity from the content related to it. This is a demanding task, so it is of great importance of how to extract the personal information in a given text. In this paper, we present an approach to extract the relevant information of the focus persons in Farsi content. Our approach contains three steps, pre-processing, attribute extraction and cross-document profile fusion. In pre-processing step, we prepare the input text as system's desired format using some existing text processing tools. In attribute extraction phase, we identify and extract all the related attributes of the target persons in a given text. To fulfill this aim, we use a pattern-matching approach. This method contains a set of keyword-based patterns and manually collected pre-compiled attribute-value candidate lists. In cross-document profile fusion phase, we integrate the distributed information of the focus persons from multiple documents and resolve ambiguity between person names. We evaluate our approach on a sample Farsi textual corpus drawn from news Websites and Wikipedia articles. The experimental results show that our approach is capable to extract personal information, thereby laying the foundation for person named-entity profiling in Farsi.
