• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.563-568
• /
• 2016

The word sense disambiguation problem is to find the correct sense of an ambiguous word having multiple senses in a dictionary in a sentence. We regard this problem as a multi-class classification problem and classify the ambiguous word by using Support Vector Machines. Context words of the ambiguous word, which are extracted from Sejong sense tagged corpus, are represented to two kinds of vector space. One vector space is composed of context words vectors having binary weights. The other vector space has vectors where the context words are mapped by word embedding model. After experiments, we acquired accuracy of 87.0% with context word vectors and 86.0% with word embedding model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.5
• /
• pp.674-677
• /
• 2020

The cultural background of users utilizing image search engines has a significant impact on the satisfaction of the search results. Therefore, it is important to analyze and understand the cultural context of images for more accurate image search. In this paper, we investigate how the cultural context of images can affect the performance of image classification. To this end, we first collected various types of images (e.g,. food, temple, etc.) with various cultural contexts (e.g., Korea, Japan, etc.) from web search engines. Afterwards, a deep transfer learning approach using VGG19 and MobileNetV2 pre-trained with ImageNet was adopted to learn the cultural features of the collected images. Through various experiments we show the performance of image classification can be differently affected according to the cultural context of images.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.115-125
• /
• 2007

The purpose of this paper is to propose a new type of NPI licensing context through French subjunctive and ne expletif. The distribution of NPIs on previous studies does not exactly correspond to negative function types. French subjunctive and ne expletif are good guidelines for reclassifying NPI licensing context. My classification is by a hierarchy of strength in negative force: overtly negative proposition > negative entailment > negative implicature. A new type of NPI licensing context is: (i) I-domain for negative implicature (ⅱ) E-domain for negative entailment and (ⅲ) overt negation.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.2
• /
• pp.81-86
• /
• 2014

In this paper, we present an effective sound classification system for recognizing the real-time context of a smartphone user. Our system avoids unnecessary consumption of limited computational resource by filtering both silence and white noise out of input sound data in the pre-processing step. It also improves the classification performance on low energy-level sounds by amplifying them as pre-processing. Moreover, for efficient learning and application of HMM classification models, our system executes the dimension reduction and discretization on the feature vectors through k-means clustering. We collected a large amount of 8 different type sound data from daily life in a university research building and then conducted experiments using them. Through these experiments, our system showed high classification performance.

• Korean Journal of Remote Sensing
• /
• v.2 no.1
• /
• pp.3-11
• /
• 1986

The current methods of estimating contest distribution function in contextual clarifier are to "classify and count", GTGM (ground-truth-guided-method) and unbiased estimator. In this paper we propose a new contextual classifier echoes context distribution is replaced by context probability that is estimated from transition probability. The classification accuracy increases considerably compared with the classical one.

• Journal of Korea Multimedia Society
• /
• v.12 no.6
• /
• pp.867-875
• /
• 2009

More and richer information sources and services are available on the web everyday. However, harmful information, such as adult content, is not appropriate for all users, notably children. Since internet is a worldwide open network, it has a limit to regulate users providing harmful contents through each countrie's national laws or systems. Additionally it is not a desirable way of developing a certain system-specific classification technology for harmful contents, because internet users can contact with them in diverse ways, for example, porn sites, harmful spams, or peer-to-peer networks, etc. Therefore, it is being emphasized to research and develop context-based core technologies for classifying harmful contents. In this paper, we propose an efficient text filter for blocking harmful texts of web documents using context-based technologies and examine which algorithms for feature selection, the process that select content terms, as features, can be useful for text categorization in all content term occurs in documents, are suitable for classifying harmful contents through implementation and experiment.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.4
• /
• pp.488-493
• /
• 2007

In general, recommendation systems quantify real-time context information obtained in the stage of collaborative filtering and use quantified context information in order to recommend services. But the recommendation systems can have problems of recommending inaccurate information because of lack of context information or classifying users into inaccurate groups because of simple classification works in the stage of quantification. In this paper, we solved the problems of lack of context information obtained in real-time by combining users' profile information used in the contents-based filtering and context information obtained in real-time. In addition, we tried collaborative filtering at the quantification stage by improving absolute classification methods to relative ones. As the result of experiments, this method improved prediction preference by 5.8% than real-time recommendation systems using context information in pure P2P environment.

• Korea Information Processing Society Review
• /
• v.11 no.6
• /
• pp.56-75
• /
• 2004

In this article we address some system characteristics and challenging issues in developing Context-aware Middleware for Ubiquitous Computing. The functionalities of a Context-aware Middleware includes gathering context data from hardware/software sensors, reasoning and inferring high-level context data, and disseminating/delivering appropriate context data to interested applications/services. The Middleware should facilitate the query, aggregation, and discovery for the contexts, as well as facilities to specify their privacy policy. Following a formal context model using ontology would enable syntactic and semantic interoperability, and knowledge sharing between different domains. Moddleware should also provide different kinds of context classification mechanical as pluggable modules, including rules written in different types of logic (first order logic, description logic, temporal/spatial logic, fuzzy logic, etc.) as well as machine-learning mechanical (supervised and unsupervised classifiers). Different mechanisms have different power, expressiveness and decidability properties, and system developers can choose the appropriate mechanism that best meets the reasoning requirements of each context. And finally, to promote the context-trigger actions in application level, it is important to provide a uniform and platform-independent interface for applications to express their need for different context data without knowing how that data is acquired. The action could involve adapting to the new environment, notifying the user, communicating with another device to exchange information, or performing any other task.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.6 no.2
• /
• pp.110-115
• /
• 2006

In pure P2P environment, it is possible to provide service by using a little real-time information without using accumulated information. But in case of using only a little information that was locally collected, quality of recommendation service can be fallen-off. Therefore, it is necessary to study a method to improve qualify of recommendation service by using users' context information. But because a great volume of users' context information can be recognized in a moment, there can be a scalability problem and there are limitations in supporting differentiated services according to fields and items. In this paper, we solved the scalability problem by clustering context information per each service field and classifying it per each user, using SOM. In addition, we could recommend proper services for users by quantifying the context information of the users belonging to the similar classification to the service requester among classified data and then using collaborative filtering.

• Journal of the Korean Society for information Management
• /
• v.29 no.2
• /
• pp.205-224
• /
• 2012

One of the limitations of BOW method is that each term is recognized only by its form, failing to represent the term's meaning or thematic background. To overcome the limitation, different profiles for each term were defined by thematic categories depending on contextual characteristics. In this study, a specific term was used as a classification feature based on its meaning or thematic background through the process of comparing the context in those profiles with the occurrences in an actual document. The experiment was conducted in three phases; term weighting, ensemble classifier implementation, and feature selection. The classification performance was enhanced in all the phases with the ensemble classifier showing the highest performance score. Also, the outcome showed that the proposed method was effective in reducing the performance bias caused by the total number of learning documents.