• The Journal of the Acoustical Society of Korea
• /
• v.26 no.1E
• /
• pp.21-26
• /
• 2007

In this paper, we propose a modified Viterbi algorithm to compensate for endpoint detection error during the decoding phase of an isolated word recognition task. Since the conventional Viterbi algorithm explores only the search space whose boundaries are fixed to the endpoints of the segmented utterance by the endpoint detector, the recognition performance is highly dependent on the accuracy level of endpoint detection. Inaccurately segmented word boundaries lead directly to recognition error. In order to relax the degradation of recognition accuracy due to endpoint detection error, we describe an unconstrained search of word boundaries and present an algorithm to explore the search space with efficiency. The proposed algorithm was evaluated by performing a variety of simulated endpoint detection error cases on an isolated word recognition task. The proposed algorithm reduced the Word Error Rate (WER) considerably, from 84.4% to 10.6%, while consuming only a little more computation power.

• MALSORI
• /
• no.59
• /
• pp.27-51
• /
• 2006

For speech recognition systems, rejection function as well as decoding function is necessary to improve the reliability. There have been many research efforts on out-of-vocabulary word rejection, however, little attention has been paid on non-target sentence rejection. Recently pronunciation approaches using speech recognition increase the need for non-target sentence rejection to provide more accurate and robust results. In this paper, we proposed filler model method and word/phoneme detection ratio method to implement non-target sentence rejection system. We made performance evaluation of filler model along to word-level, phoneme-level, and sentence-level filler models respectively. We also perform the similar experiment using word-level and phoneme-level word/phoneme detection ratio method. For the performance evaluation, the minimized average of FAR and FRR is used for comparing the effectiveness of each method along with the number of words of given sentences. From the experimental results, we got to know that word-level method outperforms the other methods, and word-level filler mode shows slightly better results than that of word detection ratio method.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.9
• /
• pp.381-388
• /
• 2014

In this paper, we deal with implementation issues of an unknown word classifier for middle-school level English writing test. We define the type of unknown words occurred in English text and discuss the detection process for unknown words. Also, we define the type of swear words occurred in students's English writings, and suggest how to handle this type of words. We implement an unknown word classifier with a swear detection module for developing an automatic English writing scoring system. By experiments with actual test data, we evaluate the accuracy of the unknown word classifier as well as the swear detection module.

• Journal of Korea Multimedia Society
• /
• v.24 no.1
• /
• pp.106-116
• /
• 2021

To increase the quality of the system in the subjective-type question grading and document classification, we need the paragraph detection. But it is not easy because it is accompanied by semantic analysis. Many researches on the paragraph detection solve the detection problem using the word based clustering method. However, the word based method can not use the order and dependency relation between words. This paper suggests the paragraph detection system using syntactic-semantic relation between words with the Korean syntactic-semantic analysis. This system is the hybrid system of word based, concept based, and syntactic-semantic tree based detection. The experiment result of the system shows it has the better result than the word based system. This system will be utilized in Korean subjective question grading and document classification.

• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.47-57
• /
• 2019

We examine the weaknesses of the existing OOXML-based MS-Word file structure, and analyze how data concealment and forgery are performed in MS-Word digital documents. In case of forgery by including hidden information in MS-Word digital document, there is no difference in opening the file with the MS-Word Processor. However, the computer system may be malfunctioned by malware or shell code hidden in the digital document. If a malicious image file or ZIP file is hidden in the document by using the structural vulnerability of the MS-Word document, it may be infected by ransomware that encrypts the entire file on the disk even if the MS-Word file is normally executed. Therefore, it is necessary to analyze forgery and alteration of digital document through internal structure analysis of MS-Word file. In this paper, we designed and implemented a mechanism to detect this efficiently and automatic detection software, and presented a method to proactively respond to attacks such as ransomware exploiting MS-Word security vulnerabilities.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.468-475
• /
• 2020

A good acoustic word embedding that can well express the characteristics of word plays an important role in wake-up word detection (WWD). However, the representation ability of acoustic word embedding may be weakened due to various types of environmental noise occurred in the place where WWD works, causing performance degradation. In this paper, we proposed triplet loss based Domain Adversarial Training (tDAT) mitigating environmental factors that can affect acoustic word embedding. Through experiments in noisy environments, we verified that the proposed method effectively improves the conventional DAT approach, and checked its scalability by combining with other method proposed for robust WWD.

• Phonetics and Speech Sciences
• /
• v.1 no.2
• /
• pp.85-94
• /
• 2009

The purpose of this study was to see the effects of pitch and length patterns on spoken word recognition. In Experiment 1, a syllable monitoring task was used to see the effects of pitch and length on the pre-lexical level of spoken word recognition. For both Seoul dialect speakers and Daegu dialect speakers, pitch and length did not affect the syllable detection processes. This result implies that there is little effect of pitch and length in pre-lexical processing. In Experiment 2, a lexical decision task was used to see the effect of pitch and length on the lexical access level of spoken word recognition. In this experiment, word frequency (low and high) as well as pitch and length was manipulated. The results showed that pitch and length information did not play an important role for Seoul dialect speakers, but that it did affect lexical decision processing for Daegu dialect speakers. Pitch and length seem to affect lexical access during the word recognition process of Daegu dialect speakers.

• Interdisciplinary Bio Central
• /
• v.3 no.4
• /
• pp.14.1-14.9
• /
• 2011

Protein homology detection is an important issue in comparative genomics. Because of the exponential growth of sequence databases, fast and efficient homology detection tools are urgently needed. Currently, for homology detection, sequence comparison methods using local alignment such as BLAST are generally used as they give a reasonable measure for sequence similarity. However, these methods have drawbacks in offering overall sequence similarity, especially in dealing with eukaryotic genomes that often contain many insertions and duplications on sequences. Also these methods do not provide the explicit models for speciation, thus it is difficult to interpret their similarity measure into homology detection. Here, we present a novel method based on Word Conservation Score (WCS) to address the current limitations of homology detection. Instead of counting each amino acid, we adopted the concept of 'Word' to compare sequences. WCS measures overall sequence similarity by comparing word contents, which is much faster than BLAST comparisons. Furthermore, evolutionary distance between homologous sequences could be measured by WCS. Therefore, we expect that sequence comparison with WCS is useful for the multiple-species-comparisons of large genomes. In the performance comparisons on protein structural classifications, our method showed a considerable improvement over BLAST. Our method found bigger micro-syntenic blocks which consist of orthologs with conserved gene order. By testing on various datasets, we showed that WCS gives faster and better overall similarity measure compared to BLAST.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.2180-2197
• /
• 2019

With the proliferation of the Android malicious applications, malware becomes more capable of hiding or confusing its malicious intent through the use of code obfuscation, which has significantly weaken the effectiveness of the conventional defense mechanisms. Therefore, in order to effectively detect unknown malicious applications on the Android platform, we propose DroidVecDeep, an Android malware detection method using deep learning technique. First, we extract various features and rank them using Mean Decrease Impurity. Second, we transform the features into compact vectors based on word2vec. Finally, we train the classifier based on deep learning model. A comprehensive experimental study on a real sample collection was performed to compare various malware detection approaches. Experimental results demonstrate that the proposed method outperforms other Android malware detection techniques.

• KIISE Transactions on Computing Practices
• /
• v.22 no.4
• /
• pp.170-177
• /
• 2016

Character encoding is a method for changing a document to a binary document file using the code table for storage in a computer. When people decode a binary document file in a computer to be read, they must know the code table applied to the file at the encoding stage in order to get the original document. Identifying the code table used for encoding the file is thus an essential part of decoding. In this paper, we propose a method for detecting the character code of the given binary document file automatically. The method uses many techniques to increase the detection rate, such as a character code range detection, escape character detection, character code characteristic detection, and commonly used word detection. The commonly used word detection method uses multiple word database, which means this method can achieve a much higher detection rate for multi-language files as compared with other methods. If the proportion of language is 20% less than in the document, the conventional method has about 50% encoding recognition. In the case of the proposed method, regardless of the proportion of language, there is up to 96% encoding recognition.