• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.6
• /
• pp.261-268
• /
• 2015

In spoken chatting systems, users'spoken queries are converted to text queries using automatic speech recognition (ASR) engines. If the top-1 results of the ASR engines are incorrect, these errors are propagated to the spoken chatting systems. To improve the top-1 accuracies of ASR engines, we propose a post-processing model to rearrange the top-n outputs of ASR engines using a ranking support vector machine (RankSVM). On the other hand, a number of chatting sentences are needed to train chatting systems. If new chatting sentences are not frequently added to training data, responses of the chatting systems will be old-fashioned soon. To resolve this problem, we propose a data collection model to automatically select chatting sentences from TV and movie scenarios using a support vector machine (SVM). In the experiments, the post-processing model showed a higher precision of 4.4% and a higher recall rate of 6.4% compared to the baseline model (without post-processing). Then, the data collection model showed the high precision of 98.95% and the recall rate of 57.14%.

• Journal of KIISE
• /
• v.44 no.6
• /
• pp.581-586
• /
• 2017

For automatic web scraping, unnecessary components such as menus and advertisements need to be removed from web pages and main contents should be extracted automatically. A content block tends to be located in the middle of a web page. In particular, Korean web documents rarely include metadata and have a complex design; a suitable method of content extraction is therefore needed. Existing content extraction algorithms use the textual and structural features of content blocks because processing visual features requires heavy computation for rendering and image processing. In this paper, we propose a new content extraction method using the tag positions in HTML as a quasi-visual feature. In addition, we develop a tag rank position, a type of tag position not affected by text length, and show that gradient boosting with the tag rank position is a very accurate content extraction method. The result of this paper shows that the content extraction method can be used to collect high-quality text data automatically from various web pages.

• The KIPS Transactions:PartB
• /
• v.13B no.2 s.105
• /
• pp.155-162
• /
• 2006

The medical thesaurus, MeSH (Medical Subject Heading), has been used as a controlled vocabulary thesaurus for English medical paper indexing for a long time. In this paper, we propose an automatic cross language keyword assignment method, which assigns English MeSH index terms to the abstract of a Korean medical paper. We compare the performance with the indexing performance of human indexers and the authors. The procedure of index term assignment is that first extracting Korean MeSH terms from text, changing these terms into the corresponding English MeSH terms, and calculating the importance of the terms to find the highest rank terms as the keywords. For the process, an effective method to solve spacing variants problem is proposed. Experiment showed that the method solved the spacing variant problem and reduced the thesaurus space by about 42%. And the experiment also showed that the performance of automatic keyword assignment is much less than that of human indexers but is as good as that of authors.

• Smart Media Journal
• /
• v.6 no.1
• /
• pp.22-29
• /
• 2017

Due to the rapid advancement and distribution of smart devices of late, document data on the Internet is on the sharp increase. The increment of information on the Web including a massive amount of documents makes it increasingly difficult for users to understand corresponding data. In order to efficiently summarize documents in the field of automated summary programs, various researches are under way. This study uses TextRank algorithm to efficiently summarize documents. TextRank algorithm expresses sentences or keywords in the form of a graph and understands the importance of sentences by using its vertices and edges to understand semantic relations between vocabulary and sentence. It extracts high-ranking keywords and based on keywords, it extracts important sentences. To extract important sentences, the algorithm first groups vocabulary. Grouping vocabulary is done using a scale of specific weight. The program sorts out sentences with higher scores on the weight scale, and based on selected sentences, it extracts important sentences to summarize the document. This study proved that this process confirmed an improved performance than summary methods shown in previous researches and that the algorithm can more efficiently summarize documents.

• CRM연구
• /
• v.2 no.2
• /
• pp.53-70
• /
• 2009

Understanding consumer behavior based on the analysis of the customer data is one essential part of analytic CRM. To do this, the analytic skills for data extraction and data processing are required to users. As a user has various kinds of questions for the consumer data analysis, the user should use database language such as SQL. However, for the firm's user, to generate SQL statements is not easy because the accuracy of the query result is hugely influenced by the knowledge of work-site operation and the firm's database. This paper proposes a natural language based database search framework finding relevant database elements. Specifically, we describe how our TableRank method can understand the user's natural query language and provide proper relations and attributes of data records to the user. Through several experiments, it is supported that the TableRank provides accurate database elements related to the user's natural query. We also show that the close distance among relations in the database represents the high data connectivity which guarantees matching with a search query from a user.

• Journal of Digital Policy
• /
• v.1 no.1
• /
• pp.1-5
• /
• 2022

The objective of this study was to examine the effectiveness of newly developed program for the recovery and relapse prevention in patients with anxiety disorder. Twenty-four patients with anxiety disorder received Mindfulness-Based Cognitive Therapy (MBCT) session weekly for a period of overall 8 weeks-program. Changes of depression, anxiety, negative and positive automatic thought were compared before and after the program. Depression and anxiety were improved significantly after the MBCT program(Z=-1.9, p=.06, Z=-2.9, p<.001). Conclusions: MBCT may be effective at reducing negative automatic thought and relieving anxiety and depressive symptoms in patients with anxiety disorder. However, large-sample, randomized controlled trials will be needed for generalization.

• International Journal of Computer Science & Network Security
• /
• v.23 no.11
• /
• pp.93-98
• /
• 2023

Instagram is one of the fastest-growing online photo social web services where users share their life images and videos with other users. Image tagging is an essential step for developing Automatic Image Annotation (AIA) methods that are based on the learning by example paradigm. Hashtags can be used on just about any social media platform, but they're most popular on Twitter and Instagram. Using hashtags is essentially a way to group together conversations or content around a certain topic, making it easy for people to find content that interests them. Practically on average, 20% of the Instagram hashtags are related to the actual visual content of the image they accompany, i.e., they are descriptive hashtags, while there are many irrelevant hashtags, i.e., stophashtags, that are used across totally different images just for gathering clicks and for search ability enhancement. Hence in this work, Sorting instagram hashtags all the way through mass tagging using HITS (Hyperlink-Induced Topic Search) algorithm is presented. The hashtags can sorted to several groups according to Jensen-Shannon divergence between any two hashtags. This approach provides an effective and consistent way for finding pairs of Instagram images and hashtags, which lead to representative and noise-free training sets for content-based image retrieval. The HITS algorithm is first used to rank the annotators in terms of their effectiveness in the crowd tagging task and then to identify the right hashtags per image.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.1
• /
• pp.27-36
• /
• 2019

In this paper, we propose a method to find the most similar answer to the user's response from the question-answer database in order to avoid generating a redundant question in retrieval-based automatic question generation system. As a question of the most similar answer to user's response may already be known to the user, the question should be removed from a set of question candidates. A similarity detector calculates a similarity between two answers by utilizing the same words, paraphrases, and sentential meanings. Paraphrases can be acquired by building a phrase table used in a statistical machine translation. A sentential meaning's similarity of two answers is calculated by an attention-based convolutional neural network. We evaluate the accuracy of the similarity detector on an evaluation set with 100 answers, and can get the 71% Mean Reciprocal Rank (MRR) score.

• Journal of Information Technology Applications and Management
• /
• v.27 no.1
• /
• pp.49-58
• /
• 2020

Lately, a large amount of textual data have been poured out of the Internet and the technology to refine them is needed. Most of these data are long text and often have no title. Therefore, in this paper, we propose a technique to combine the sequence-to-sequence model of RNN and the REINFORCE algorithm to generate the title of the long text automatically. In addition, the TextRank algorithm was applied to extract a summarized text to minimize information loss in order to protect the shortcomings of the sequence-to-sequence model in which an information is lost when long texts are used. Through the experiment, the techniques proposed in this study are shown to be superior to the existing ones.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.26 no.4
• /
• pp.263-279
• /
• 2022

This paper presents an automatic inspection of defects in semiconductor images. We devise a statistical method to find defects on homogeneous background from the observation that it has a log-normal distribution. If computer aided design (CAD) data is available, we use it to construct a signed distance function (SDF) and change the pixel values so that the average of pixel values along the level curve of the SDF is zero, so that the image has a homogeneous background. In the absence of CAD data, we devise a hybrid method consisting of a model-based algorithm and two neural networks. The model-based algorithm uses the first right singular vector to determine whether the image has a linear or complex structure. For an image with a linear structure, we remove the structure using the rank 1 approximation so that it has a homogeneous background. An image with a complex structure is inspected by two neural networks. We provide results of numerical experiments for the proposed methods.