Taekeun Park

A forwarding scheme for reliable and energy-efficient data delivery in cluster-based sensor networks

This letter proposes a novel forwarding scheme for reliable and energy-efficient data delivery in the cluster-based sensor networks. While multiple nodes in a cluster receive a packet, only one node among them is elected to send the acknowledgement back and then to broadcast it to the next cluster. With the binary exponential backoff algorithm for the election, the proposed scheme is more reliable and energy-efficient than existing forwarding schemes for the cluster-based sensor networks.

Appropriate SACK Handling for Temporary Disconnections in a Wireless Multi-Homed Environment

In this letter, we describe inappropriate SACK (Selective Acknowledgement) handling in SCTP multi-homed environment, and propose a simple modification. The main idea of the modification is that a sender does not increase the congestion control window size of a primary path when availability of the path is ambiguous. Through simulation, we show that the performance is much improved with this simple modification compared to standard SCTP.

A contents recommendation scheme using user's affection and shopping motive

This paper proposes a contents recommendation scheme using users affection and shopping motive to improve the accuracy of recommendation. In this paper, joy, sadness, anger, happiness, and relaxation are considered as the users affection. And the users shopping motive is classified into four types: low-utility/low-pleasure, low-utility/high-pleasure, highutility/low-pleasure, and high-utility/high-pleasure. experimental results show that the proposed scheme improves the accuracy of contents recommendation compared to the ecommendation without considering users affection and shopping motive.

A method for automatically translating print books into electronic Braille books

In this paper, a method for automatically translating scanned images from print books into electronic Braille books is proposed with the objective of reducing the amount of time and cost required for producing Braille books. The proposed method consists of processes for identifying character and image areas in a scanned image, automatically translating characters and images into Braille and tactile graphics, respectively, and positioning Braille and tactile graphics into an electronic Braille page. Experimental results show that the proposed method drastically reduces the time required to translate a print book into an electronic Braille book. Despite the drastic reduction in translation time, the method proposed in this paper does not compromise the ability to recognize information for the visually impaired compared to manually produced Braille books, demonstrating its feasibility in practical applications. Therefore, the proposed method is expected to significantly reduce the time and cost required for producing Braille books, and provide more reading materials for the visually impaired, making significant contributions to enhancing their knowledge and welfare.

A Character Identification Method using Postpositions for Animate Nouns in Korean Novels

Submitted:June 20, 2016 1 st Revision:July 20, 2016 Accepted:July 26, 2016 * 본 연구는 문화체육관광부 및 한국콘텐츠진흥원의 2016년도 문화기술 연구개발 지원사업으로 수행되었음. ** 단국대학교 응용컴퓨터공학과 교수, 교신저자 ***단국대학교 응용컴퓨터공학과 교수 Novels includes various character names, depending on the genre and the spatio-temporal background of the novels and the nationality of characters. Besides, characters and their names in a novel are created by the author’s pen and imagination. As a result, any proper noun dictionary cannot include all kind of character names which have been created or will be created by authors. In addition, since Korean does not have capitalization feature, character names in Korean are harder to detect than those in English. Fortunately, however, Korean has postpositions, such as “-ege” and “hante”, used by a sentient being or an animate object (noun). We call such postpositions as animate postpositions in this paper. In a previous study, the authors manually selected character names by referencing both Wikipedia and well-known people dictionaries after utilizing Korean morpheme analyzer, a proper noun dictionary, postpositions (e.g., “-ga”, “-eun”, “-neun”, “-eui”, and “-ege”), and titles (e.g., “buin”), in order to extract social networks from three novels translated into or written in Korean. But, the precision, recall, and F-measure rates of character identification are not presented in the study. In this paper, we evaluate the quantitative contribution of animate postpositions to character identification from novels, in terms of precision, recall, and F-measure. The results show that utilizing animate postpositions is a valuable and powerful tool in character identification without a proper noun dictionary from novels translated into or written in Korean. Keyword:Information Extraction, Korean Novels, Character Identification, Postpositions for Animate Nouns, Korean Linguistic Feature 韓國IT서비스學會誌 第15卷 第3號 2016年 9月, pp.115-125 116 Taekeun Park.Seung-Hoon Kim

Novel Character Identification Utilizing Semantic Relation with Animate Nouns in Korean

For identifying speakers of quoted speech or extracting social networks from literature, it is indispensable to extract character names and nominals. However, detecting proper nouns in the novels translated into or written in Korean is harder than in English because Korean does not have a capitalization feature. In addition, it is almost impossible for any proper noun dictionary to include all kinds of character names that have been created or will be created by authors. Fortunately, a previous study shows that utilizing postpositions for animate nouns is a simple and effective tool for character identification in Korean novels without a proper noun dictionary and a training corpus. In this article, we propose a character identification method utilizing the semantic relation with known animate nouns. For 80 novels in Korean, the proposed method increases the micro- and macro-average recall by 13.68% and 11.86%, respectively, while decreasing the micro-average precision by 0.28% and increasing the macro-average precision by 0.07% compared to the previous study. If we focus on characters that are responsible for more than 1% of the character name mentions in each novel, the micro- and macro-average F-measure of the proposed method are 96.98% and 97.32%, respectively.

A Contents Recommendation Scheme Based on Collaborative Filtering Using Consumer's Affection and Consumption Type

ABSTRACT Collaborative filtering is a popular technique used for the recommendation system, but its performance, especially the accuracy of recommendation, depends on how to define the reference group. This paper proposes a new contents recommendation scheme based on collaborative filtering technique whose reference groups are created by consumers affection and consumption type in order to improve the accuracy of recommendation. In this paper, joy, sadness, anger, happiness, and relax are considered as the consumers affection. And, low-utility / low-pleasure, low-utility / high-pleasure, high-utility / low-pleasure, and high-utility / high-pleasure are considered as the consumers shopping types. Experimental results show that the proposed scheme improves the accuracy of recommendation compared to the recommendation scheme considering neither consumers affection nor consumption type.Key Words:Contents Recommendation, Affection, Consumption Type, Collaborative Filtering

Efficient and fast resource reservation by the maximum localization of QoS re-establishment

To overcome the mobility impact on RSVP, many schemes have been proposed based on Mobile IP regional registration and passive reservation in advance. Although the regional registration and in advance reservation reduce the QoS interruption time, the latter may demand intolerable bandwidth. This letter introduces a novel approach to reduce the QoS interruption time by maximizing the localization of QoS re-establishment in the regional registration environment. The proposed scheme identifies the exact path segment affected by mobility. The QoS interruption time of the proposed scheme is comparable to its low bound without in advance reservation.