김민호 교수

Songs feel emotionally different to listeners depending on their lyrical contents, even when melodies are similar. Accordingly, when using features related to melody, like tempo, rhythm, tune, and musical note, it is difficult to classify emotions accurately through the existing music emotion classification methods. This paper therefore proposes a method for lyrics-based emotion classification using feature selection by partial syntactic analysis. Based on the existing emotion ontology, four kin

We sincerely thank our shepherd Dr. Ranveer Chandra and the anonymous reviewers for their valuable feedback. This work was supported by Samsung Research Funding Center for Future Technology under Project Number SRFC-IT1402-01.

The user authentication is an important part of network security. Several strong-password authentication protocols have been introduced, but a secure scheme, which probably withstands to several known attacks, is not yet available. Recently, a hash-based strong-password authentication scheme was described in [2], which withstands to the several attacks, including replay, passwordflle compromise, denial-of-service, and insider attacks. However, we show that this protocol is still vulnerable to st

Error words that appear in Korean texts can be largely categorized into non-word spelling errors and context-sensitive spelling errors. Of the two, context-sensitive spelling errors are shown only when considering the meaning of the word in the given context and its syntactic relation, and they are the most difficult to correct among spelling errors. Context-sensitive spelling errors can be categorized into homophone errors, typographical errors, grammatical errors, and cross-word boundary error

Abstract We focus on open‐domain question‐answering tasks that involve a chain‐of‐reasoning, which are primarily implemented using large language models. With an emphasis on cost‐effectiveness, we designed EffiChainQA , an architecture centered on the use of small language models. We employed a retrieval‐based language model to address the limitations of large language models, such as the hallucination issue and the lack of updated knowledge. To enhance reasoning capabilities, we introduced a qu

The goal of word sense disambiguation(WSD) is to determine which sense of an ambiguous word is invoked in a particular use of the word. It plays an important role in many natural language processing applications such as machine translation and information retrieval. This paper proposes a method for automatic word sense disambiguation based on Korean WordNet. The basic assumption of this suggested method is that contextual words provide strong and consistent clues to the sense of an ambiguous wor

This research, which considered context-sensitive spelling error correction as the classification problem of words along with the context as the solution of semantic ambiguity, has a limitation that correction can be corrected in a particular words pair. To overcome this, this research suggested a technique for detecting and correcting context-sensitive spelling error probabilistically, by selecting the whole eojeol1 as the target words and generating corresponding available candidate. The conte

Real-time, plausible visual and haptic feedback of deformable objects without shape artifacts is important in surgical simulation environments to avoid distracting the user. We propose to leverage highly parallel stream processing, available on the newest generation graphics cards, to increase the level of both visual and haptic fidelity. We implemented this as part of the University of Florida's haptic surgical authoring kit.

Context-sensitive spelling-error correction methods are largely classified into rule-based methods and statistical data-based methods, the latter of which is often preferred in research. Statistical error correction methods consider context-sensitive spelling error problems as word-sense disambiguation problems. The method divides a vocabulary pair, for correction, which consists of a correction target vocabulary and a replacement candidate vocabulary, according to the context. The present paper

Supervised disambiguation using a large amount of corpus data delivers better performance than other word sense disambiguation methods. However, it is not easy to construct large-scale, sense-tagged corpora since this requires high cost and time. On the other hand, implementing unsupervised disambiguation is relatively easy, although most of the efforts have not been satisfactory. A primary reason for the performance degradation of unsupervised disambiguation is that the semantic occurrence prob

Secure communication is an important issue in networks and user authentication is a very important part of the security. Several strong-password authentication protocols have been introduced, but there is no fully secure authentication scheme that can resist all known attacks. We propose enhanced secure schemes with registration and login protocols, and add the “forget password ” and password/verifier change protocols. We show that our scheme is more secure against guessing, stolen-verifier, rep

Fire spread models (FSMs) are used to reproduce fire behavior and can simulate fire propagation over landscapes. As wildfires have emerged into a global phenomenon with far-reaching impacts on the natural and built environments, FSM simulations provide crucial information to better understand and predict fire behavior in various landscapes. In this study, we tested Cell2Fire, a recently developed cellular automata-based FSM, against benchmarking models used in the U.S., Canada, and Chile. We exp

Novikov and Kiselev [7] proposed an authentication method of a user from a remote autonomous object. Recently, Yang et al. [12] and Awasthi [1] have pointed out that the Novikov-Kiselev scheme is insecure against the man-in-the-middle attack. In this article, we propose an improved version of the Novikov-Kiselev scheme to overcome such vulnerability.