• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.4
• /
• pp.138-147
• /
• 2007

Metabolic pathway is a series of chemical reactions occuning within a cell and can be used for drug development and understanding of life phenomenon. Many biologists are trying to extract metabolic pathway information from huge literatures for their metabolic-circuit regulation study. We propose a text-mining technique based on the keyword and pattern. Proposed technique utilizes a web robot to collect huge papers and stores them into a local database. We use gene ontology to increase compound recognition rate and NCBI Tokenizer library to recognize useful information without compound destruction. Furthermore, we obtain useful sentence patterns representing metabolic pathway from papers and KEGG database. We have extracted 66 patterns in 20,000 documents for Glycosphingolipid species from KEGG, a representative metabolic database. We verify our system for nineteen compounds in Glycosphingolipid species. The result shows that the recall is 95.1%, the precision 96.3%, and the processing time 15 seconds. Proposed text mining system is expected to be used for metabolic pathway reconstruction.

• Korean Journal of Biological Psychiatry
• /
• v.27 no.1
• /
• pp.18-26
• /
• 2020

Objectives The aim was to find effective vectorization and classification models to predict a psychiatric diagnosis from text-based medical records. Methods Electronic medical records (n = 494) of present illness were collected retrospectively in inpatient admission notes with three diagnoses of major depressive disorder, type 1 bipolar disorder, and schizophrenia. Data were split into 400 training data and 94 independent validation data. Data were vectorized by two different models such as term frequency-inverse document frequency (TF-IDF) and Doc2vec. Machine learning models for classification including stochastic gradient descent, logistic regression, support vector classification, and deep learning (DL) were applied to predict three psychiatric diagnoses. Five-fold cross-validation was used to find an effective model. Metrics such as accuracy, precision, recall, and F1-score were measured for comparison between the models. Results Five-fold cross-validation in training data showed DL model with Doc2vec was the most effective model to predict the diagnosis (accuracy = 0.87, F1-score = 0.87). However, these metrics have been reduced in independent test data set with final working DL models (accuracy = 0.79, F1-score = 0.79), while the model of logistic regression and support vector machine with Doc2vec showed slightly better performance (accuracy = 0.80, F1-score = 0.80) than the DL models with Doc2vec and others with TF-IDF. Conclusions The current results suggest that the vectorization may have more impact on the performance of classification than the machine learning model. However, data set had a number of limitations including small sample size, imbalance among the category, and its generalizability. With this regard, the need for research with multi-sites and large samples is suggested to improve the machine learning models.

• The KIPS Transactions:PartB
• /
• v.11B no.2
• /
• pp.227-232
• /
• 2004

When we extract a representative term from text for information retrieval system or a special information for information retrieval and text milling system, we often need to solve the anaphora resolution problem. The antecedent decision problem of a pronoun is one of the major issues for anaphora resolution. In this paper, we are suggesting a method of deciding an antecedent of the third personal pronouns, such as “he/she/they” to analyze the contents of documents precisely. Generally, the antecedent of the third personal Pronouns seem to be the subject of the current statement or previous statement, and also it occasionally happens more than twice. Based on these characteristics, we have found rules for deciding an antecedent, by investigating a case of being an antecedent from the personal pronouns, which appears in the current statement and the previous statements. Since the heuristic rule differs on the case of the third personal pronouns, we described it as subjective case, objective case, and possessive case based on the case of the pronouns. We collected 300 sentences that include a pronoun from the newspaper articles on political issues. The result of our experiment shows that the recall and precision ratio on deciding the antecedent of the third personal pronouns are 79.0% and 86.8%, respectively.

• Journal of Intelligence and Information Systems
• /
• v.18 no.1
• /
• pp.91-107
• /
• 2012

As the Internet becomes more popular, many people use it to communicate. With the increasing number of personal homepages, blogs, and social network services, people often expose their personal information online. Although the necessity of those services cannot be denied, we should be concerned about the negative aspects such as personal information leakage. Because it is impossible to review all of the past records posted by all of the people, an automatic personal information detection method is strongly required. This study proposes a method to detect or classify online documents that contain personal information by analyzing features that are common to personal information related documents and learning that information based on the Na$\ddot{i}$ve Bayes algorithm. To select the document classification algorithm, the Na$\ddot{i}$ve Bayes classification algorithm was compared with the Vector Space classification algorithm. The result showed that Na$\ddot{i}$ve Bayes reveals more excellent precision, recall, F-measure, and accuracy than Vector Space does. However, the measurement level of the Na$\ddot{i}$ve Bayes classification algorithm is still insufficient to apply to the real world. Lewis, a learning algorithm researcher, states that it is important to improve the quality of category features while applying learning algorithms to some specific domain. He proposes a way to incrementally add features that are dependent on related documents and in a step-wise manner. In another experiment, the algorithm learns the additional dependent features thereby reducing the noise of the features. As a result, the latter experiment shows better performance in terms of measurement than the former experiment does.

• Journal of KIISE:Databases
• /
• v.34 no.3
• /
• pp.193-205
• /
• 2007

Increasing amount of digital images requires more accurate and faster way of image retrieval. So far, image retrieval method includes content-based retrieval and keyword based retrieval, the former utilizing visual features such as color and brightness and the latter utilizing keywords which describe the image. However, the effectiveness of these methods as to providing the exact images the user wanted has been under question. Hence, many researchers have been working on relevance feedback, a process in which responses from the user are given as a feedback during the retrieval session in order to define user’s need and provide improved result. Yet, the methods which have employed relevance feedback also have drawbacks since several feedbacks are necessary to have appropriate result and the feedback information can not be reused. In this paper, a novel retrieval model has been proposed which annotates an image with a keyword and modifies the confidence level of the keyword in response to the user’s feedback. In the proposed model, not only the images which have received positive feedback but also the other images with the visual features similar to the features used to distinguish the positive image are subjected to confidence modification. This enables modifying large amount of images with only a few feedbacks ultimately leading to faster and more accurate retrieval result. An experiment has been performed to verify the effectiveness of the proposed model and the result has demonstrated rapid increase in recall and precision while receiving the same number of feedbacks.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.7
• /
• pp.221-228
• /
• 2020

Part-of-Speech (POS) tagged corpus is a collection of electronic text in which each word is annotated with a tag as the corresponding POS and is widely used for various training data for natural language processing. The training data generally assumes that there are no errors, but in reality they include various types of errors, which cause performance degradation of systems trained using the data. To alleviate this problem, we propose a novel method for detecting errors in the existing POS tagged corpus using the classifier of XGBoost and cross-validation as evaluation techniques. We first train a classifier of a POS tagger using the POS-tagged corpus with some errors and then detect errors from the POS-tagged corpus using cross-validation, but the classifier cannot detect errors because there is no training data for detecting POS tagged errors. We thus detect errors by comparing the outputs (probabilities of POS) of the classifier, adjusting hyperparameters. The hyperparameters is estimated by a small scale error-tagged corpus, in which text is sampled from a POS-tagged corpus and which is marked up POS errors by experts. In this paper, we use recall and precision as evaluation metrics which are widely used in information retrieval. We have shown that the proposed method is valid by comparing two distributions of the sample (the error-tagged corpus) and the population (the POS-tagged corpus) because all detected errors cannot be checked. In the near future, we will apply the proposed method to a dependency tree-tagged corpus and a semantic role tagged corpus.

• The KIPS Transactions:PartB
• /
• v.17B no.3
• /
• pp.255-262
• /
• 2010

When a system is developed, requirements document is generated by requirement analysts and then translated to formal specifications by specifiers. If a formal specification can be generated automatically from a natural language requirements document, system development cost and system fault from experts' misunderstanding will be decreased. A pronoun can be classified in personal and demonstrative pronoun. In the characteristics of requirements document, the personal pronouns are almost not occurred, so we focused on the decision of antecedent for a demonstrative pronoun. For the higher accuracy in analysis of requirements document automatically, finding antecedent of demonstrative pronoun is very important for elicitation of formal requirements automatically from natural language requirements document via natural language processing. The final goal of this research is to automatically generate formal specifications from natural language requirements document. For this, this paper, based on previous research [3], proposes an anaphora resolution system to decide antecedent of pronoun using natural language processing from natural language requirements document in Korean. This paper proposes heuristic rules for the system implementation. By experiments, we got 92.45%, 69.98% as recall and precision respectively with ten requirements documents.

• Journal of the Korea Convergence Society
• /
• v.12 no.9
• /
• pp.11-19
• /
• 2021

A software system is required to change during its life cycle due to various requirements such as adding functionalities, fixing bugs, and adjusting to new computing environments. Such program code modification should be considered as carefully as a new system development becase unexpected software errors could be introduced. In addition, when reusing open source programs, we can expect higher quality software if code changes of the open source program are predicted in advance. This paper proposes a Convolutional Neural Network (CNN)-based deep learning model to predict source code changes. In this paper, the prediction of code changes is considered as a kind of a binary classification problem in deep learning and labeled datasets are used for supervised learning. Java projects and code change logs are collected from GitHub for training and testing datasets. Software metrics are computed from the collected Java source code and they are used as input data for the proposed model to detect code changes. The performance of the proposed model has been measured by using evaluation metrics such as precision, recall, F1-score, and accuracy. The experimental results show the proposed CNN model has achieved 95% in terms of F1-Score and outperformed the multilayer percept-based DNN model whose F1-Score is 92%.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.359-366
• /
• 2021

A number of construction companies in Korea invest considerable human and financial resources to construct a system for managing apartment defect data and for categorizing repair tasks. Thus, this study proposes machine learning models to automatically classify defect complaint text-data into one of the sub categories of 'finishing work' (i.e., one of the defect repair tasks). In the proposed models, we employed two word representation methods (Bag-of-words, Term Frequency-Inverse Document Frequency (TF-IDF)) and two machine learning classifiers (Support Vector Machine, Random Forest). In particular, we conducted both binary- and multi- classification tasks to classify 9 sub categories of finishing work: home appliance installation work, paperwork, painting work, plastering work, interior masonry work, plaster finishing work, indoor furniture installation work, kitchen facility installation work, and tiling work. The machine learning classifiers using the TF-IDF representation method and Random Forest classification achieved more than 90% accuracy, precision, recall, and F1 score. We shed light on the possibility of constructing automated defect classification systems based on the proposed machine learning models.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.25 no.4
• /
• pp.65-74
• /
• 2021

In order to efficiently inspect rapidly increasing old tunnels in many well-developed countries, many inspection methodologies have been proposed using imaging equipment and image processing. However, most of the existing methodologies evaluated their performance on a clean concrete surface with a limited area where other objects do not exist. Therefore, this paper proposes a 6-step framework for tunnel crack detection deep learning model development. The proposed method is mainly based on negative sample (non-crack object) training and Cascade Mask R-CNN. The proposed framework consists of six steps: searching for cracks in images captured from real tunnels, labeling cracks in pixel level, training a deep learning model, collecting non-crack objects, retraining the deep learning model with the collected non-crack objects, and constructing final training dataset. To implement the proposed framework, Cascade Mask R-CNN, an instance segmentation model, was trained with 1561 general crack images and 206 non-crack images. In order to examine the applicability of the trained model to the real-world tunnel crack detection, field testing is conducted on tunnel spans with a length of about 200m where electric wires and lights are prevalent. In the experimental result, the trained model showed 99% precision and 92% recall, which shows the excellent field applicability of the proposed framework.