• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.2
• /
• pp.69-76
• /
• 2018

Referring expressions are natural language constructions used to identify particular objects within a scene. In this paper, we propose a new deep neural network model for referring expression comprehension. The proposed model finds out the region of the referred object in the given image by making use of the rich information about the referred object itself, the context object, and the relationship with the context object mentioned in the referring expression. In the proposed model, the object matching score and the relationship matching score are combined to compute the fitness score of each candidate region according to the structure of the referring expression sentence. Therefore, the proposed model consists of four different sub-networks: Language Representation Network(LRN), Object Matching Network (OMN), Relationship Matching Network(RMN), and Weighted Composition Network(WCN). We demonstrate that our model achieves state-of-the-art results for comprehension on three referring expression datasets.

• Journal of Digital Contents Society
• /
• v.8 no.1
• /
• pp.93-106
• /
• 2007

Web Applications are rapidly increasing and the structure becomes very complicated. However, when users explore such complex Web applications, they cannot often grasp the current location and get the information that they want. Therefore, a novel approach to model the navigation of Web application contents is required. In this study, a framework has been presented for modeling adaptive context-based contents navigation of Web applications. The framework performs activities including navigation analysis, navigation design, and navigation realization. first, in navigation analysis domain is analyzed by using use case, focusing on navigation. Next, in navigation design three models have been produced: a navigation information model, a profile, and a navigation interface model. Finally, in navigation realization a Webpage navigation model and a component navigation model have been produced. In this work, several formal definitions and rules for checking validity of navigation model have also been provided.

• Educational Technology International
• /
• v.11 no.1
• /
• pp.171-192
• /
• 2010

This study was to investigate the elaborated project based learning model for scientifically gifted in the context of R & E project learning. It is important for the scientifically gifted to provide the appropriate learning environments instead of general learning model for the gifted. Although R & E project learning model is effective, the model has the limitations of managing the course for the scientifically gifted. To improve R & E learning model, the elaborated project based learning model was suggested with integration of both project based learning model and goal based scenario. The elaborated project-based learning model was comprised with 'basic learning process', 'elaboration through inquiry', and 'presentation and reflection'. To measure the satisfaction, eighty scientifically gifted students participated in the class. The result shows that learners were satisfied with the elaborated project-based learning up to 90%, and teachers were satisfied with this model up to 77%.

• Communications of Mathematical Education
• /
• v.27 no.3
• /
• pp.179-203
• /
• 2013

The purpose of this study was to discuss the example that developed geometry model textbook based on storytelling using real-life context. To achieve this purpose, we first elaborated the meaning of the textbook based on storytelling with real-life context, and then we discussed the outline of the story and the summary of each lesson. This study defined the storytelling textbook with real-life context as the textbook consisting of activities that explored and organized mathematical concepts by using real-life situations as materials of stories. The geometry textbook we developed employed two real-life materials, a map and a set square: we used a map for the coordinate geometry and a set square for the equation of a line. To attract students' interest, we introduced confrontation between a teacher and two students and a villain. We implemented experimentation with the textbook based on storytelling in order to verify its validity. The participants were 25 students that were enrolled in a high school in Seoul. Among them, 17 participants were surveyed. Students' answers from the survey questionnaire suggested that the geometry textbook we developed based on storytelling helped them learn mathematics and that the instruments such as a map and a set square helped them understand mathematical concepts. However, their opinion implied that the story of the textbook needed to be improved so that the story reflected more realistic contexts that were familiar with students.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.5
• /
• pp.470-476
• /
• 2014

A point of interest is a specific point location such as a cafe, a gallery, a shop, or a park. It consists of a name, a category, a location, and so on. Its information is necessary for location-based application, above all category is basic information. However, category information should be automatically gathered because it costs high to gather it manually. In this paper, we propose a novel method to estimate category of POIs automatically using an inner word and local context. An inner word is a word that contains POI's name. Their name sometimes expose category information. Thus, their name is used as inner word information in estimating category of POIs. Local context information means words around a POI's name in a document that mentioned the name. The context include information to estimate category. The evaluation of the proposed method is performed on two data sets. According to the experimental results, proposed model using combination inner word and local context show higher accuracy than that of model using each.

• Journal of KIISE
• /
• v.42 no.5
• /
• pp.629-636
• /
• 2015

The types of errors corrected by a Korean spelling and grammar checker can be classified into isolated-term spelling errors and context-sensitive spelling errors (CSSE). CSSEs are difficult to detect and to correct, since they are correct words when examined alone. Thus, they can be corrected only by considering the semantic and syntactic relations to their context. CSSEs, which are frequently made even by expert wiriters, significantly affect the reliability of spelling and grammar checkers. An existing Korean spelling and grammar checker developed by P University (KSGC 4.5) adopts hand-made correction rules for correcting CSSEs. The KSGC 4.5 is designed to obtain very high precision, which results in an extremely low recall. Our overall goal of previous works was to improve the recall without considerably lowering the precision, by generalizing CSSE correction rules that mainly depend on linguistic knowledge. A variety of rule-based methods has been proposed in previous works, and the best performance showed 95.19% of average precision and 37.56% of recall. This study thus proposes a statistics based method using a conditional probability model with dynamic window sizes. in order to further improve the recall. The proposed method obtained 97.23% of average precision and 50.50% of recall.

• Journal of Information Technology Services
• /
• v.12 no.4
• /
• pp.399-422
• /
• 2013

The aim of this study is to find the model that best explains members' continued use intention in virtual communities by comparing technical and social perspectives applied in IS context. With the eight major variables identified from prior studies, four alternative models were formulated:1) A base model from Information Systems (IS) continuance perspective, 2) IS continuance model with technology-acceptance perspective, 3) IS continuance model with social-capital perspective, and 4) IS continuance model with socio-technical combined perspective. The adequacies of these four models with different perspective highlighted were tested using survey data collected from virtual community users in Korea. The findings indicated that the IS continuance model and social-capital perspective is the most efficient model that best explains the members' continued use intention in virtual communities.

• Architectural research
• /
• v.12 no.2
• /
• pp.15-22
• /
• 2010

Our prediction model is based on the development of "Semantic Location Model." It embodies geometrical and topological information which can increase the efficiency in prediction and make it easy to manipulate the prediction model. Data mining is being implemented to extract the inhabitant's location patterns generated day by day. As a result, the self-learning system will be able to semantically predict the inhabitant's location in advance. This context-aware system brings about the key component of the ubiquitous computing environment. First, we explain the semantic location model and data mining methods. Then the location prediction model for the ubiquitous computing system is described in details. Finally, the prototype system is introduced to demonstrate and evaluate our prediction model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.1
• /
• pp.25-31
• /
• 2014

Korean grammar checkers typically detect context-dependent errors by employing heuristic rules; these rules are formulated by language experts and consisted of lexical items. Such grammar checkers, unfortunately, show low recall which is detection ratio of errors in the document. In order to resolve this shortcoming, a new error-decision rule-generalization method that utilizes the existing KorLex thesaurus, the Korean version of Princeton WordNet, is proposed. The method extracts noun classes from KorLex and generalizes error-decision rules from them using the Tree Cut Model and information-theory-based MDL (minimum description length).

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.538-561
• /
• 2020

Recommender Systems (RecSys) have a major role in e-commerce for recommending products, which they may like for every user and thus improve their business aspects. Although many types of RecSyss are there in the research field, the state of the art RecSys has focused on finding the user similarity based on sequence (e.g. purchase history, movie-watching history) analyzing and prediction techniques like Recurrent Neural Network in Deep learning. That is RecSys has considered as a sequence prediction problem. However, evaluation of similarities among the customers is challenging while considering temporal aspects, context and multi-component ratings of the item-records in the customer sequences. For addressing this issue, we are proposing a Deep Learning based model which learns customer similarity directly from the sequence to sequence similarity as well as item to item similarity by considering all features of the item, contexts, and rating components using Dynamic Temporal Warping(DTW) distance measure for dynamic temporal matching and 2D-GRU (Two Dimensional-Gated Recurrent Unit) architecture. This will overcome the limitation of non-linearity in the time dimension while measuring the similarity, and the find patterns more accurately and speedily from temporal and spatial contexts. Experiment on the real world movie data set LDOS-CoMoDa demonstrates the efficacy and promising utility of the proposed personalized RecSys architecture.