• Proceedings of the KSPS conference
• /
• 2002.11a
• /
• pp.183-186
• /
• 2002

The speech detection is one of the important problems in real-time speech recognition. The accurate detection of speech boundaries is crucial to the performance of speech recognizer. In this paper, we propose a speech detector based on Mel-band selection through training. In order to show the excellence of the proposed algorithm, we compare it with a conventional one, so called, EPD-VAA (EndPoint Detector based on Voice Activity Detection). The proposed speech detector is trained in order to better extract keyword speech than other speech. EPD-VAA usually works well in high SNR but it doesn't work well any more in low SNR. But the proposed algorithm pre-selects useful bands through keyword training and decides the speech boundary according to the energy level of the sub-bands that is previously selected. The experimental result shows that the proposed algorithm outperforms the EPD-VAA.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.3
• /
• pp.51-56
• /
• 2009

In a language, noun and keyword extraction is a key element in information processing. When it comes to processing Korean language information, however, there are still a lot of problems with noun and keyword extraction. This paper proposes an effective noun extraction method that considers noun emergence features. The proposed method can be effectively used in areas like information retrieval where large volumes of documents and data need to be processed in a fast manner. In this paper, a category-based keyword construction method is also presented that uses an unsupervised learning technique to ensure high volumes of queries are automatically classified. Our experimental results show that the proposed method outperformed both the supervised learning-based X2 method known to excel in keyword extraction and the DF method, in terms o classification precision.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.5
• /
• pp.670-673
• /
• 2020

Recent advances in artificial intelligence have allowed for easier sentiment analysis (e.g. positive or negative forecast) of documents such as a finance reports. In this paper, we investigate a method to apply text mining techniques to extract in the financial report using deep learning, and propose an accounting model for the effects of sentiment values in financial information. For sentiment analysis with keyword detection in the financial report, we suggest the input layer with extracted keywords, hidden layers by learned weights, and the output layer in terms of sentiment scores. Our approaches can help more effective strategy for potential investors as a professional guideline using sentiment values.

• MALSORI
• /
• no.43
• /
• pp.151-161
• /
• 2002

This study presents performance analysis results of speaker adaptation for keyword spotting system. In this paper, we implemented MLLR (Maximum Likelihood Linear Regression) method on our middle size vocabulary keyword spotting system. This system was developed for directory services of universities and colleges. The experimental results show that speaker adaptation reduces the false alarm rate to 1/3 with the preservation of the mis-detection ratio. This improvement is achieved when speaker adaptation is applied to not only keyword models but also non-keyword models.

• The Journal of Society for e-Business Studies
• /
• v.14 no.4
• /
• pp.59-73
• /
• 2009

Keyword extraction is an important and essential technique for text mining applications such as information retrieval, text categorization, summarization and topic detection. A set of keywords extracted from a large-scale electronic document data are used for significant features for text mining algorithms and they contribute to improve the performance of document browsing, topic detection, and automated text classification. This paper presents a keyword extraction technique that can be used to detect topics for each news domain from a large document collection of internet news portal sites. Basically, we have used six variants of traditional TF-IDF weighting model. On top of the TF-IDF model, we propose a word filtering technique called 'cross-domain comparison filtering'. To prove effectiveness of our method, we have analyzed usefulness of keywords extracted from Korean news articles and have presented changes of the keywords over time of each news domain.

• The Journal of the Korea Contents Association
• /
• v.18 no.7
• /
• pp.449-458
• /
• 2018

As the usage of social network services increases, event information occurring offline is spreading more rapidly. Therefore, studies have been conducted to detect events by analyzing social data. In this paper, we propose a graph based event detection scheme considering user interest in social networks. The proposed scheme constructs a keyword graph by analyzing tweets posted by users. We calculates the interest measure from users' social activities and uses it to identify events by considering changes in interest. Therefore, it is possible to eliminate events that are repeatedly posted without meaning and improve the reliability of the results. We conduct various performance evaluations to demonstrate the superiority of the proposed event detection scheme.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.01a
• /
• pp.111-113
• /
• 2016

빠른 속도로 쏟아지고 있는 각종 발행물, 논문들에 대한 표절 검토는 표절 검출 알고리즘을 통해 직접적인 복제, 짜깁기, 말 바꾸어 쓰기 등을 검토하거나 표절 검토자가 직접 해당 문서의 키워드를 검색하여 확인하는 방식으로 이루어지고 있다. 하지만 점점 더 늘어나는 방대한 양의 문서들에 대한 표절 검토 작업은 더욱 정교한 검토 방법론을 필요로 하고 있으며, 이를 돕기 위해 문서의 직접적인 단어나 복제 비교에서 더 나아가 문서의 내용을 비교하여 비슷한 내용의 문서들을 필터링 및 검출할 수 있는 방법을 제안한다. 문서의 내용을 비교하기 위해 키워드 추출 알고리즘을 선행하며, 이를 통해 문서의 핵심 내용을 비교할 수 있는 기반을 마련하여 표절 검토자의 작업의 정확성과 속도를 향상시키고자 한다.

• The Korean Journal of Applied Statistics
• /
• v.29 no.4
• /
• pp.719-728
• /
• 2016

In this paper, we analyzed IoT patent data using the social network analysis of keyword community network in patents related to Internet of Things technology. To identify the difference of IoT patent trends between Korea and USA, 100 Korea patents and 100 USA patents were collected, respectively. First, we first extracted important keywords from IoT patent abstracts using the TF-IDF weight and their correlation and then constructed the keyword network based on the selected keywords. Second, we constructed a keyword community network based on the keyword community and performed social network analysis. Our experimental results showed while Korea patents focus on the core technologies of IoT (such as security, semiconductors and image process areas), USA patents focus on the applications of IoT (such as the smart home, interactive media and telecommunications).

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.26 no.2
• /
• pp.397-404
• /
• 2016

Illegal and harmful contents on the Internet has been an issue and been increased in Korea. They are often posted on the billboard and website of small enterprise and government office. Those illegal and harmful contents can relate to crime and suspicious activity, so, we need a detection system. However, to date the detection itself has been conducted manually by a person. In this paper, we develop an automated URL detection scheme for detecting a drug trafficking by using Google. This system works by analyzing the frequently used keywords in a drug trafficking and generate a keyword dictionary to store words for future search. The suspected drug trafficking URL are automatically collected based on the keyword dictionary by using Google search engine. The suspicious URL can be detected by classifying and numbering each domain from the collection of the suspected URL. This proposed automated URL detection can be an effective solution for detecting a drug trafficking, also reducing time and effort consumed by human-based URL detection.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.7
• /
• pp.632-640
• /
• 2002

In this paper, we proposed ACM (Anti-filler confidence measure) to compensate shortcoming of conventional RLJ-CM (RLJ-CM) and NCM (normalized CM), and integrated proposed ACM and conventional NCM using HCM (hybrid CM). Proposed ACM analyzes that FA (false acceptance) happens by the construction method of anti-phone model, and presumed phoneme sequence in actuality using phoneme recognizer to compensate this. We defined this as anti-phone model and used in confidence measure calculation. Analyzing feature of two confidences measure, conventional NCM shows good performance to FR (false rejection) and proposed ACM shows good performance in FA. This shows that feature of each other are complementary. Use these feature, we integrated two confidence measures using weighting vector α And defined this as HCM. In MDR (missed detection rate) 10% neighborhood, HCM is 0.219 FA/KW/HR (false alarm/keyword/hour). This is that Performance improves 22% than used conventional NCM individually.