• Journal of KIISE:Software and Applications
• /
• v.27 no.1
• /
• pp.84-95
• /
• 2000

In this paper, we propose an integrated indexing method with compound noun segmentation and noun phrase synthesis. Statistical information is used in the compound noun segmentation and natural language processing techniques are carefully utilized in the noun phrase synthesis. Firstly, we choose index terms from simple words through morphological analysis and part-of-speech tagging results. Secondly, noun phrases are automatically synthesized from the syntactic analysis results. If syntactic analysis fails, only morphological analysis and tagging results are applied. Thirdly, we select compound nouns from the tagging results and then segment and re-synthesize them using statistical information. In this way, segmented and synthesized terms are used together as index terms to supplement the single terms. We demonstrate the effectiveness of the proposed integrated indexing method for Korean compound noun processing using KTSET2.0 and KRIST SET which are a standard test collection for Korean information retrieval.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.6
• /
• pp.529-536
• /
• 2004

The main idea of this paper is to combine position information in sentence and query type classification to make the documents ranking to query more accessible. First, the use of conceptual graphs for the representation of document contents In information retrieval is discussed. The method is based on well-known strategies of text comparison, such as Dice Coefficient, with position-based weighted term. Second, we introduce a method for learning query type classification that improves the ability to retrieve answers to questions from Question Answering system. Proposed methods employ naive bayes classification in machine learning fields. And, we used a collection of approximately 30,000 question-answer pairs for training, obtained from Frequently Asked Question(FAQ) files on various subjects. The evaluation on a set of queries from international TREC-9 question answering track shows that the method with machine learning outperforms the underline other systems in TREC-9 (0.29 for mean reciprocal rank and 55.1% for precision).

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.8 no.1
• /
• pp.50-59
• /
• 2007

In this paper, an anchor shot detection method, which is a basic technology for managing news videos for index and retrieval purposes is proposed. To do that, two most popular news program such as 'KBS 9 Hour News' and 'MBC 9 Hour News' are analyzed and 4-step rule based detection method is proposed First, in the preprocessing, video shot boundaries are detected and the 1st frame of each shot is extracted as a key frame. Then, the detected shot is declared as an anchor shot, if all the following 4 conditions are satisfied. 1) There is an anchor face in the key frame of a shot. 2) Spatial distribution of edges in the key frame is adequate. 3) Background color information of the key frame is similar to the color information of an anchor model. 4) Motion rate in the shot is low. In order to show the validity of the proposed method, three 'KBS 9 Hour News' and three 'MBC 9 Hour News', which have total running time of 108 in minute and are broadcasted at different days, are used for experiments. Average detection rates showed 0.97 in precision, 1.0 in recall, and 0.98 in F-measure.

• Journal of KIISE:Information Networking
• /
• v.30 no.5
• /
• pp.614-620
• /
• 2003

As the proliferation of Internet, mobile agent related technologies are examined for possible growth and evolution. In information retrieval, network and distributed system management, and electronic commerce, mobile agent is more flexible than the traditional communication paradigm. Despite the performance benefits, mobile agent is not used widely in the market because it is very vulnerable to a variety of attacks. In many studies related the security vulnerability for a mobile agent, the high security causes the performance to degrade. In this paper, we propose and evaluate the efficient security model for mobile agent using TDGM(Trusted Domain & Guide Manager), which provides three kinds of services : the trusted domain management, the security service and the travel plan guide. The result clearly shows that this model provides high security and minimizes the performance degradation.

• Journal of KIISE
• /
• v.43 no.1
• /
• pp.47-53
• /
• 2016

While the Internet develops rapidly, a huge amount of image data collected from smart phones, digital cameras and black boxes are being shared through social media sites. Generally, social images are handled by tagging them with information. Due to the ease of sharing multimedia and the explosive increase in the amount of tag information, it may be considered too much hassle by some users to put the tags on images. Image retrieval is likely to be less accurate when tags are absent or mislabeled. In this paper, we suggest a method of extracting tags from social images by using image content. In this method, CNN(Convolutional Neural Network) is trained using ImageNet images with labels in the training set, and it extracts labels from instagram images. We use the extracted labels for automatic image tagging. The experimental results show that the accuracy is higher than that of instagram retrievals.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.7
• /
• pp.1621-1630
• /
• 2009

There is a lot of recent research to provide services of dynamic distributed contents in P2P-based distributed environment. Distributed services of contents objects, however, have problems in QoS and dynamic management of segments. This paper proposed a new segment management method for the service of P2P-based distributed contents. The proposed method manages groups by the SGM(Segment Group Manager). The SGM manages streaming efficiently by the grouping of segments to be served in P2P environment. The segments in the same group cooperate for improved QoS using the management structure based on the distance and relationship. The distance-based management structure is for the improvement of retrieval efficiency while the relationship-based management structure is for the improvement of service ratio. The simulation results of the proposed method showed improvements in average transmission efficiency and average service rate. The improvement is 8% - 30% in average transmission efficiency and 10% - 30% in average service rate.

• Journal of KIISE:Databases
• /
• v.29 no.2
• /
• pp.128-137
• /
• 2002

In this paper, we propose the LPC+-file for efficient indexing of high-dimensional image data. With the proliferation of multimedia data, there Is an increasing need to support the indexing and retrieval of high-dimensional image data. Recently, the LPC-file (5) that based on vector approximation has been developed for indexing high-dimensional data. The LPC-file gives good performance especially when the dataset is uniformly distributed. However, compared with for the uniformly distributed dataset, its performance degrades when the dataset is clustered. We improve the performance of the LPC-file for the strongly clustered image dataset. The basic idea is to adaptively partition the data space to find subspaces with high-density clusters and to assign more bits to them than others to increase the discriminatory power of the approximation of vectors. The total number of bits used to represent vector approximations is rather less than that of the LPC-file since the partitioned cells in the LPC+-file share the bits. An empirical evaluation shows that the LPC+-file results in significant performance improvements for real image data sets which are strongly clustered.

• Journal of KIISE:Databases
• /
• v.32 no.1
• /
• pp.24-42
• /
• 2005

Characteristics of biological data including genome sequences are heterogeneous and various. Although the need of management systems for genome sequencing which should reflect biological characteristics has been raised, most current biological databases provide restricted function as repositories for biological data. Therefore, this paper describes a management system of nucleotide sequences at the level of biological laboratories. It includes format transformation, editing, storing and retrieval for collected nucleotide sequences from public databases, and handles sequence produced by experiments. It uses BSML based on XML as a common format in order to extract data fields and transfer heterogeneous sequence formats. To manage sequences and their changes, version management system for originated DNA is required so as to detect transformed new sequencing appearance and trigger database update. Our experimental results show that applying active trigger rules to manage changes of sequences can automatically store changes of sequences into databases.

• Journal of KIISE:Databases
• /
• v.32 no.1
• /
• pp.71-85
• /
• 2005

Various researches on moving object modeling, uncertainty processing, and moving object indexing have been tarried out in the field of moving object databases. However. previous location tracking systems cannot efficiently retrieve location data of vehicles, because they manage all location information of vehicles using the conventional database. In this paper, we design the vehicle location management systen that is able to manage and retrieve vehicle locations efficiently in mobile environment. The proposed system consists of a server for managing vehicle locations and mobile clients. The system is able to not only process spatiotemporal queries related to locations of moving vehicles but also Provide moving vehicles' locations which are not stored in the system. The system is also able to manage vehicle location data effectively using a moving object index.

• Journal of KIISE:Databases
• /
• v.29 no.2
• /
• pp.119-127
• /
• 2002

Video server which stores video streams should serve more users with the requested streams rapidly, satisfying their time constraints. However, the number of users supported and the response times for the requested services in video server are in the oppositional relationship with viewpoint of disk throughput. In this paper, we propose new policy that can provide not only rapid responses for the service requests but also increase the number of users supported in the storage system of video server. The policy admits the random movements of disk heads for urgent service requests but maintains the sequential or restricted movements of disk heads and eliminates rewind time of disk heads for less-urgent service requests in storage system of multiple disks environment so that it can not satisfy only the QoS of users but also support more users.