• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2E
• /
• pp.51-55
• /
• 2004

In this paper, we propose a style-specific language model adaptation scheme using n-gram based tf*idf similarity for Korean spontaneous speech recognition. Korean spontaneous speech shows especially different style-specific characteristics such as filled pauses, word omission, and contraction, which are related to function words and depend on preceding or following words. To reflect these style-specific characteristics and overcome insufficient data for training language model, we estimate in-domain dependent n-gram model by relevance weighting of out-of-domain text data according to their n-. gram based tf*idf similarity, in which in-domain language model include disfluency model. Recognition results show that n-gram based tf*idf similarity weighting effectively reflects style difference.

• The Journal of the Acoustical Society of Korea
• /
• v.36 no.6
• /
• pp.436-441
• /
• 2017

Computing music similarity is indispensable in constructing music retrieval system. This paper focuses on the cover song search among various music-retrieval tasks. We investigate the cover song search method based on the chroma n-gram to reduce storage for feature DB and enhance search accuracy. Specifically we propose t-tab n-gram, n-gram selection method, and n-gram set comparison method. Experiments on the widely used music dataset confirmed that the proposed method improves cover song search accuracy as well as reduces feature storage.

• MALSORI
• /
• no.46
• /
• pp.117-126
• /
• 2003

This paper presents our language model adaptation for Korean spontaneous speech recognition. Korean spontaneous speech is observed various characteristics of content and style such as filled pauses, word omission, and contraction as compared with the written text corpus. Our approaches focus on improving the estimation of domain-dependent n-gram models by relevance weighting out-of-domain text data, where style is represented by n-gram based tf/sup */idf similarity. In addition to relevance weighting, we use disfluencies as Predictor to the neighboring words. The best result reduces 9.7% word error rate relatively and shows that n-gram based relevance weighting reflects style difference greatly and disfluencies are good predictor also.

• KIISE Transactions on Computing Practices
• /
• v.20 no.12
• /
• pp.711-716
• /
• 2014

With the gradual increase of interest in plagiarism and intelligent file content search, the demand for similarity measuring between two sentences is increasing. There is a lot of researches for sentence similarity measurement methods in various directions such as n-gram, edit-distance and LSA. However, these methods have their own advantages and disadvantages. In this paper, we propose a new sentence similarity measurement method approaching from another direction. The proposed method uses the set-based POI data search that improves search performance compared to the existing hard matching method when data includes the inverse, omission, insertion and revision of characters. Using this method, we are able to measure the similarity between two sentences more accurately and more quickly. We modified the data loading and text search algorithm of the set-based POI data search. We also added a word operation algorithm and a similarity measure between two sentences expressed as a percentage. From the experimental results, we observe that our sentence similarity measurement method shows better performance than n-gram and the set-based POI data search.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.402-404
• /
• 1998

최근 영상 정보를 저장하는 시스템의 급증으로, 영상의 특징 요소들의 유사성(similarity)에 근거하여 영상을 분류.검색하는 기술에 많은 관심을 보이고 있다. 본 논문에서는 칼라영상의 분류를 위해 기존의 N$\times$M-grams를 변형한 Color N$\times$M-grams를 제안한다. Color N$\times$M-grams는 영상의 칼라정보를 이용하여 영상고유의 구조 정보를 추출한 후 유사성을 측정하여 영상을 분류한다. 제안된 방법의 성능 평가를 위하여 39쌍의 Benchmark 영상을 사용하여 실험하였다. 실험결과, 제안한 Color N$\times$M-grams를 사용한 방법이 기존의 N$\times$M-grams를 사용하여 칼라 영상을 분류하는 방법보다 1순위로 분류되는 비율에 있어서 약 19% 더 좋은 결과를 보였다.

• Proceedings of the Korea Contents Association Conference
• /
• 2007.11a
• /
• pp.12-16
• /
• 2007

The function of a protein is closely co-related with its subcellular location(s). Given a protein sequence, therefore, how to determine its subcellular location is a vitally important problem. We have developed a new prediction method for protein subcellular location(s), which is based on n-gram feature extraction and k-nearest neighbor (kNN) classification algorithm. It classifies a protein sequence to one or more subcellular compartments based on the locations of top k sequences which show the highest similarity weights against the input sequence. The similarity weight is a kind of similarity measure which is determined by comparing n-gram features between two sequences. Currently our method extract penta-grams as features of protein sequences, computes scores of the potential localization site(s) using kNN algorithm, and finally presents the locations and their associated scores. We constructed a large-scale data set of protein sequences with known subcellular locations from the SWISS-PROT database. This data set contains 51,885 entries with one or more known subcellular locations. Our method show very high prediction precision of about 93% for this data set, and compared with other method, it also showed comparable prediction improvement for a test collection used in a previous work.

• Informatization Policy
• /
• v.26 no.4
• /
• pp.85-106
• /
• 2019

Recently, there have been increasing cases of collective petitions filed in the electronic petitions system. However, there is no efficient management system, raising concerns on side effects such as increased administrative workload and mass production of social conflicts. Aimed at suggesting a methodology for estimating electronic collective petitions using anomaly detection and corpus linguistics-based content analysis, this study conducted the followings: i) a theoretical review of the concept of collective petitions, ii) estimation of electronic collective petitions using anomaly detection based on nonparametric unsupervised learning, iii) a content similarity analysis on petitions using n-gram cosine angle distance, and iv) a case study on the Voice of Citizens of Changwon City, through which the utility of the proposed methodology, policy implications and future tasks were reviewed.

• Proceedings of the KSPS conference
• /
• 2003.05a
• /
• pp.83-86
• /
• 2003

This paper presents our style-based language model adaptation for Korean conversational speech recognition. Korean conversational speech is observed various characteristics of content and style such as filled pauses, word omission, and contraction as compared with the written text corpora. For style-based language model adaptation, we report two approaches. Our approaches focus on improving the estimation of domain-dependent n-gram models by relevance weighting out-of-domain text data, where style is represented by n-gram based tf*idf similarity. In addition to relevance weighting, we use disfluencies as predictor to the neighboring words. The best result reduces 6.5％ word error rate absolutely and shows that n-gram based relevance weighting reflects style difference greatly and disfluencies are good predictor.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.5
• /
• pp.1141-1151
• /
• 2018

In an APT attack, the communication stage between infected hosts and C&C(Command and Control) server is the key stage for intrusion into the attack target. Attackers can control multiple infected hosts by the C&C Server and direct intrusion and exploitation. If the C&C Server is exposed at this stage, the attack will fail. Therefore, in recent years, the Domain Generation Algorithm (DGA) has replaced DNS in C&C Server with a short time interval for making detection difficult. In particular, it is very difficult to verify and detect all the newly registered DNS more than 5 million times a day. To solve these problems, this paper proposes a model to judge DGA-DNS detection by the morphological similarity analysis of normal DNS and DGA-DNS, and to determine the sign of APT attack through it, then we verify its validity.

• Journal of Software Assessment and Valuation
• /
• v.12 no.2
• /
• pp.1-8
• /
• 2016

In this paper, we study abstracting and identifying license file from a package to prevent unintentional intellectual property infringement because of lost/modified/confliction of license information when redistributing open source software. To invest character of the license files, we analyzed 322 licenses by n-gram and TF-IDF methods, and abstract license files from the packages. We identified license information with a similarity of the registered licenses by cosine measurement.