• Journal of the Korea Society of Computer and Information
• /
• v.26 no.3
• /
• pp.1-7
• /
• 2021

In this paper, optimized media data retrieval and transmission based on the Hybrid CDN/P2P architecture and selective storage through user's prediction of requestability enable seamless data transfer to users and reduction of unnecessary traffic. We also propose a new media management method to minimize the possibility of transmission delay and packet loss so that media can be utilized in real time. To this end, we construct each media into logical segments, continuously compute weights for each segment, and determine whether to store segment data based on the calculated weights. We also designate scattered computing nodes on the network as local groups by distance and ensure that storage space is efficiently shared and utilized within those groups. Experiments conducted to verify the efficiency of the proposed technique have shown that the proposed method yields a relatively good performance evaluation compared to the existing methods, which can enable both initial latency reduction and seamless transmission.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.11
• /
• pp.4084-4104
• /
• 2021

Visual place recognition (VPR) is a fundamental yet challenging task of mobile robot navigation and localization. The existing VPR methods are usually based on some pairwise similarity of image descriptors, so they are sensitive to visual appearance change and also computationally expensive. This paper proposes a simple yet effective four-step method that achieves adaptive convolutional neural network (CNN) landmark matching for VPR. First, based on the features extracted from existing CNN models, the regions with higher significance scores are selected as landmarks. Then, according to the coordinate positions of potential landmarks, landmark matching is improved by removing mismatched landmark pairs. Finally, considering the significance scores obtained in the first step, robust image retrieval is performed based on adaptive landmark matching, and it gives more weight to the landmark matching pairs with higher significance scores. To verify the efficiency and robustness of the proposed method, evaluations are conducted on standard benchmark datasets. The experimental results indicate that the proposed method reduces the feature representation space of place images by more than 75% with negligible loss in recognition precision. Also, it achieves a fast matching speed in similarity calculation, satisfying the real-time requirement.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.1
• /
• pp.35-40
• /
• 2022

In this paper, we propose a method of constructing clinical knowledge with clinical concept relations and predicting diseases based on a deep learning model to support clinical decision-making. Clinical terms in UMLS(Unified Medical Language System) and cancer-related medical knowledge are classified into five categories. Medical related documents in Wikipedia are extracted using the classified clinical terms. Clinical concept relations are established by matching the extracted medical related documents with the extracted clinical terms. After deep learning using clinical knowledge, a disease is predicted based on medical terms expressed in a query. Thereafter, medical terms related to the predicted disease are selected as an extended query for clinical document retrieval. To validate our method, we have experimented on TREC Clinical Decision Support (CDS) and TREC Precision Medicine (PM) test collections.

• Journal of Internet of Things and Convergence
• /
• v.9 no.1
• /
• pp.19-24
• /
• 2023

Recently, with the recent rapid development of memory technology, various types of memory are developed and are used to improve processing speed in data management systems. In particular, NAND flash memory is used as a main media for storing data in memory-based storage devices because it has a nonvolatile characteristic that it can maintain data even at the power off state. However, since the recently studied memory-based storage device consists of various types of memory such as MRAM and PRAM as well as NAND flash memory, research on memory management technology is needed to improve data processing performance and efficiency of media in a storage system composed of different types of memories. In this paper, we propose a memory mapping scheme thought technique for efficiently managing data in the storage device composed of various memories for data management. The proposed idea is a method of managing different memories using a single mapping table. This method can unify the address scheme of data and reduce the search cost of data stored in different memories for data tiering.

• Journal of Korea Society of Industrial Information Systems
• /
• v.14 no.5
• /
• pp.11-18
• /
• 2009

Ajax based on Java Script receives attention as an alternative to ActiveX technology. Most portal sites in korea show a tendency to reopen existing services by combining the technology, because it supports most web browsers, and has the advantages of such a brilliant interface, excellent speed, and traffic reduction through asynchronous interaction. This paper modeled and implemented a multilingual meta-search service using the Ajax and open APIs provided by international famous sites. First, a Korean query is translated into one of the language of 54 countries around the world by Google translation API, and then the translated result is used to search the information of the social web sites such as Flickr, Youtube, Daum, and Naver. Searched results are displayed fast by dynamic loading of portion of the screen using Ajax. Our system can reduce server traffic and per-packet communications charges by preventing redundant transmission of unnecessary information.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.11
• /
• pp.21-30
• /
• 2009

Semantic web uses ontology languages such as RDF, RDFS, and OWL to define the metadata on the web. There have been many researching efforts in the semantic web technologies based on an agent for extracting triple and relation about concept of ontology. But the extraction of relation and triple about the concept of ontology based on an agent ends up writing a limited query statement as characteristics of an agent. As for this, there is the less of flexibility when extracting triple and relation about the other concept of ontology. We are need a query tool for flexible information retrieval of ontology that is can access the standard ontology and can be used standard query language. In this paper, we propose a SPARQL query tool that is can access the OWL ontology via HTTP protocol and it can be used to make a query. Query result can be output to the soap message. These operations can be support the web service.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.10
• /
• pp.31-41
• /
• 2009

In the Semantic Web, it is possible to provide intelligent information retrieval and automated web services by defining a concept of information resource and representing a semantic relation between resources with meta data and ontology. It is very important to manage semantic data such as ontology and meta data efficiently for implementing essential functions of the Semantic Web. Thus we propose an index structure to support more accurate search results and efficient query processing by considering semantic and structural features of the semantic data. Especially we use a graph data model to express semantic and structural features of the semantic data and process various type of queries by using graph model based path expressions. In this paper the proposed index aims to distinguish our approach from earlier studies and involve the concept of the Semantic Web in its entirety by querying on primarily extracted structural path information and secondary extracted one through semantic inferences with ontology. In the experiments, we show that our approach is more accurate and efficient than the previous approaches and can be applicable to various applications in the Semantic Web.

• Journal of Digital Convergence
• /
• v.20 no.1
• /
• pp.179-184
• /
• 2022

With the development of information and communication technology, the production and distribution of digital contents is rapidly increasing, and the distribution of illegally copied contents also increases, causing various problems. In order to prevent illegal distribution of contents, a DRM (Digital Rights Management)-based approach can be used, but in a situation where the contents are already copied and distributed, a method of searching and detecting the duplicated contents is required. In this paper, a duplication detection method based on the contents of video content is proposed. The proposed method divides the video into scene units using the visual rhythm extracted from the video, and hierarchically applies the playback time and color feature values of each divided scene to quickly and efficiently detect duplicate videos in a large database. Through experiments, it was shown that the proposed method can reliably detect various replication modifications.

• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.13-21
• /
• 2022

Recent deep learning-based research shows excellent performance in most natural language processing (NLP) fields with pre-trained language models. In particular, the auto-encoder-based language model proves its excellent performance and usefulness in various fields of Korean language understanding. However, the decoder-based Korean generative model even suffers from generating simple sentences. Also, there is few detailed research and data for the field of conversation where generative models are most commonly utilized. Therefore, this paper constructs multi-turn dialogue data for a Korean generative model. In addition, we compare and analyze the performance by improving the dialogue ability of the generative model through transfer learning. In addition, we propose a method of supplementing the insufficient dialogue generation ability of the model by extracting recommended response candidates from external knowledge information through a retrival model.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.75-77
• /
• 2021

A variety of search engines are currently in use. The search engine supports the retrieval of data required by users through three stages: crawling, index generation, and output of search results based on meta-tag information. However, a large number of documents obtained by searching for keywords are often unrelated or scarce. Because of these problems, it takes time and effort to grasp the content from the search results and classify the accuracy. The index of search engines is updated periodically, but the criteria for weighted values and update periods are different from one search engine to another. Therefore, this paper uses the LSTM model, which extracts the relationship between keywords entered by the user and documents instead of the existing search engine, and improves the relationship between keywords and documents by entering keywords that the user wants to find.