• Journal of KIISE
• /
• v.44 no.7
• /
• pp.686-691
• /
• 2017

Telemarketing has become the center of marketing action of the industry in the information society. Recently, machine learning has emerged in many areas, especially, financial prediction. Financial data consists of lots of unlabeled data in most parts, and therefore, it is difficult for humans to perform their labeling. In this paper, we propose a fusion method of semi-supervised learning for automatic labeling of unlabeled data to predict telemarketing. Specifically, we integrate labeling results of label propagation and co-training with a decision tree. The data with lower reliabilities are removed, and the data are extracted that have consistent label from two labeling methods. After adding them to the training set, a decision tree is learned with all of them. To confirm the usefulness of the proposed method, we conduct the experiments with a real telemarketing dataset in a Portugal bank. Accuracy of the proposed method is 83.39%, which is 1.82% higher than that of the conventional method, and precision of the proposed method is 19.37%, which is 2.67% higher than that of the conventional method. As a result, we have shown that the proposed method has a better performance as assessed by the t-test.

• Korean Journal of Community Nutrition
• /
• v.27 no.2
• /
• pp.132-145
• /
• 2022

Objectives: This study examined trends in the health status and dietary behavior changes by region using the raw data from the 2008 ~ 2019 Community Health Survey. Methods: This study analyzed the data of 2,738,572 people among the raw data of the Community Health Survey from 2008 to 2019. The regional differences in health status and dietary behavior were examined by classifying the regions into capital and non-capital regions, and the non-capital regions were classified into metropolitan cities and provinces. A chi-square test was conducted on the body mass index (BMI), diagnosis of diabetes and hypertension, frequency of eating breakfast, salty taste in usual diet, recognition of nutrition labeling, reading of nutrition labeling, and utilization of nutrition labeling. Results: In determining obesity using the BMI, the normal weight by year decreased, and the obesity rate by year was 34.6% in 2019, which increased by 13% compared to 2008. In addition, the diabetes diagnosis rate and hypertension diagnosis rate continued to increase with the year. Both diabetes and hypertension diagnosis rates were higher in the non-capital regions than in the capital region. Eating breakfast five to seven times per week was most common and showed a significant decreasing trend by year (P < 0.001). The percentage of respondents who said they eat slightly bland foods increased from 19.5% in 2008 to 19.9% in 2010 and then to 22.1% in 2013. The percentage then decreased to 19.9% in 2019, but showed an overall increasing trend (P < 0.001). According to the region, the capital region had a higher percentage than the non-capital region. The nutrition labeling's recognition rate and utilization rate increased yearly, whereas the reading rate decreased. Conclusions: The study results presented the primary data necessary to develop nutrition education programs and establish strategies for local nutrition management projects to improve disease prevention and dietary problems.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.5
• /
• pp.502-510
• /
• 2024

In this paper, we proposed the weakly-supervised deep learning algorithm for active sonar target recognition based on pseudo labeling using Conventional Recurrent Neural Network (CRNN) model widely used for acoustic signal processing because it can effectively utilize small and unbalanced active sonar data. Active sonar simulation data assuming two different SNRs and clutter environments were used in the training and testing process, and spectrogram obtained by applying Short Time Fourier Transform (STFT) to the simulation data was used as a feature factor for algorithm training. The algorithm proposed in this paper was evaluated based on the target and nontarget F1-score using test data independent of training data. As a result, it was confirmed that the CRNN model showed significant performance not only in typical acoustic signal processing but also active sonar target recognition. Also, pseudo-labeling helps to improve the performance of the active sonar target recognition algorithm used the CRNN model.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.10
• /
• pp.31-38
• /
• 2015

In this paper, we propose the ensemble-based classification model which extracts just new data patterns from the streaming-data by using clustering and generates new classification models to be added to the ensemble in order to reduce the number of data labeling while it keeps the accuracy of the existing system. The proposed technique performs clustering of similar patterned data from streaming data. It performs the data labeling to each cluster at the point when a certain amount of data has been gathered. The proposed technique applies the K-NN technique to the classification model unit in order to keep the accuracy of the existing system while it uses a small amount of data. The proposed technique is efficient as using about 3% less data comparing with the existing technique as shown the simulation results for benchmarks, thereby using clustering.

• ETRI Journal
• /
• v.36 no.3
• /
• pp.429-438
• /
• 2014

Semantic role labeling (SRL) is a task in natural-language processing with the aim of detecting predicates in the text, choosing their correct senses, identifying their associated arguments, and predicting the semantic roles of the arguments. Developing a high-performance SRL system for a domain requires manually annotated training data of large size in the same domain. However, such SRL training data of sufficient size is available only for a few domains. Constructing SRL training data for a new domain is very expensive. Therefore, domain adaptation in SRL can be regarded as an important problem. In this paper, we show that domain adaptation for SRL systems can achieve state-of-the-art performance when based on structural learning and exploiting a prior model approach. We provide experimental results with three different target domains showing that our method is effective even if training data of small size is available for the target domains. According to experimentations, our proposed method outperforms those of other research works by about 2% to 5% in F-score.

• Korean Journal of Organic Agriculture
• /
• v.28 no.2
• /
• pp.139-154
• /
• 2020

The purpose of this study is to improve some problems of the country-of-origin labeling based on the perception of consumers. For this, we surveyed 636 people. The questions of the survey are largely divided into three categories; 1) criteria and subject for imposition of fine, 2) the possibility of getting consumers confused with the products using domestic regional names as domestic products, 3) criteria for the country-of-origin transplantation of agricultural products and forestry products. According to the results, more than 30.0% of consumers preferred that it is adequate for imposing fine as much as its total sales, regardless of the type of business. In addition, in the case of products using domestic regional names, consumers can be confused about the products with domestic ones, even though there is a standard for confusing country-of-origin labeling. Standard for changing the country-of-origin of agricultural, forestry products and livestock, fisheries products are not balanced. The results of this study can be used as basis data for revising the country-of-origin labeling.

• Asian Journal of Business Environment
• /
• v.14 no.3
• /
• pp.13-21
• /
• 2024

Purpose: This study explores the potential use of food e-labels for restaurants to solve the current inadequacies in food labeling within the restaurant sector. Additionally, the study examines the feasibility and scalability of implementing e-labels for food labeling purposes, investigates consumers' perceptions of e-labels for restaurant offerings, and assesses the value of implementing e-labels. Research design, data and methodology: The value of food e-labels was estimated using the contingent valuation method. Samples were selected from the survey, considering the distribution of population, using stratified sampling method. In the survey, respondents were provided with information explaining the food e-label and were asked whether they would accept the proposed amount for food e-labeling. Results: Estimation results revealed that the individual demographic factors of the respondents significantly influenced their willingness to pay (WTP), along with their food purchasing behavior and the degree of food labeling checking. Based on the estimated results, WTP was calculated to be 2,624 KRW. Conclusions: The study findings can serve as a reference for related businesses and policies, suggesting the need for further research and detailed discussions. To activate food e-labeling, promotion and education are essential complements to mere regulatory implementation.

• Survey Research
• /
• v.11 no.3
• /
• pp.57-79
• /
• 2010

This study evaluated the policy performance of i) Public Notice of Critical Information, ii) Substantiation of Facts in Labeling and Advertising, iii) Temporary Injunctions, and iv) Advertisement Correcting Misrepresented Facts, which were main policies belonged to Fair Labeling and Advertising Act(hereinafter referred to as "FLA Act"). The data was collected by visiting 76 persons personally, who were consumer policy and law experts, labeling and advertising staffs of corporations, and persons in charge of policies including public officials dealing with consumer policies at Korea Fair Trade Commission, while using a structured questionnaire at the same time. The survey was performed to examine the general policy performance and evaluation the results of FLA Act by evaluation methods. The results of the analysis are comprehensively summarized as follows. There were differences in the ranking of policies evaluated by labeling and advertising staffs of corporations and persons in charge of policies according to evaluation methods, and, in Simple Evaluation, higher scores were gained compared to Weighted Evaluation which reflected weighted values or Fuzzy Evaluation. The result shows that evaluation results can vary in policy performance evaluation according to evaluation methods.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.38 no.12
• /
• pp.1801-1810
• /
• 2009

This study was carried out to investigate food-nutrition labeling perception and utilization classified by age in Busan. The survey was conducted from March 26 to April 30, 2008 by questionnaires and data analyzed by SPSS program. The results are summarized as follows: reasons for purchase of the processed food was 'delicious' in elementary school children and middle & high school students, but was 'easy to eat and cook' in the adults groups (p<0.001). The criteria for choice of the processed foods was 'taste' in all of the subjects. Eighty seven point five percent of the over 60's do not know about food labeling and 70.1% of them did not check the food label. The first confirmed items for buying the processed foods was 'expiration date' in all of the subjects (71.1%). In elementary school children, middle & high school students, 20's & 30's group, the ratio of awareness of nutrition label was higher than the 40's & 50's and over 60's group. For reading of nutrition label, all of the subjects except elementary group replied 'often' (p<0.001). For the experience of education and publicity on food-nutrition labeling, 54.3% of the subjects replied 'often', and there was a significant difference by age. For the necessity of education and publicity on food-nutrition labeling, 49.5% of the subjects replied 'necessary'. There was significant positive correlation between degree of checking of nutrition label and degree of checking of food label, accuracy of knowledge of processed food, necessity of education and publicity. Therefore, education and publicity on food-nutrition labeling for the subjects are required to encourage them to choose more nutritious food and have healthier dietary pattern.

• Journal of KIISE:Databases
• /
• v.33 no.1
• /
• pp.129-137
• /
• 2006

An XML labeling scheme is an efficient encoding method to determine the ancestor-descendant relationships of elements and the orders of siblings. Recently, many dynamic XML documents have appeared in the Web Services and the AXML(the Active XML), so we need to manage them with a dynamic XML labeling scheme. The prime number labeling scheme is a representative scheme which supports dynamic XML documents. It determines the ancestor-descendant relationships between two elements with the feature of prime numbers. When a new element is inserted into the XML document using this scheme, it has an advantage that an assigning the label of new element don't change the label values of existing nodes. But it has to have additional expensive operations and data structure for maintaining the orders of siblings. In this paper, we suggest the order number sharing method and algorithms categorized by the insertion positions of new nodes. They greatly minimize the existing method's sibling order maintenance cost.