• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.4
• /
• pp.93-106
• /
• 2020

This research explores how imported automobile companies can develop their strategies to improve the outcome of their recalls. For this, the researchers analyzed patterns of recall demand, classified recall types based on the demand patterns and examined response strategies, considering plans on how to procure parts and induce customers to visit workshops, recall execution capacity and costs. As a result, recalls are classified into four types: U-type, reverse U-type, L- type and reverse L-type. Also, as determinants of the types, the following factors are further categorized into four types and 12 sub-types of recalls: the height of maximum demand, which indicates the volatility of recall demand; the number of peaks, which are the patterns of demand variations; and the tail length of the demand curve, which indicates the speed of recalls. The classification resulted in the following: L-type, or customer-driven recall, is the most common type of recalls, taking up 25 out of the total 36 cases, followed by five U-type, four reverse L-type, and two reverse U-type cases. Prior studies show that the types of recalls are determined by factors influencing recall execution rates: severity, the number of cars to be recalled, recall execution rate, government policies, time since model launch, and recall costs, etc. As a component demand forecast model for automobile recalls, this study estimated the ARIMA model. ARIMA models were shown in three models: ARIMA (1,0,0), ARIMA (0,0,1) and ARIMA (0,0,0). These all three ARIMA models appear to be significant for all recall patterns, indicating that the ARIMA model is very valid as a predictive model for car recall patterns. Based on the classification of recall types, we drew some strategic implications for recall response according to types of recalls. The conclusion section of this research suggests the implications for several aspects: how to improve the recall outcome (execution rate), customer satisfaction, brand image, recall costs, and response to the regulatory authority.

• Journal of Sensor Science and Technology
• /
• v.33 no.3
• /
• pp.119-124
• /
• 2024

Effective fire sensing is important to protect lives and property from the disaster. In this paper, we present an intelligent visual sensing method for detecting fires based on machine learning techniques. The proposed method involves a two-step process. In the first step, fire and non-fire images are used to train a convolutional neural network (CNN), and in the next step, feature vectors consisting of 256 values obtained from the CNN are used for the learning of a support vector machine (SVM). Linear and nonlinear SVMs with different parameters are intensively tested. We found that the proposed hybrid method using an SVM with a linear kernel effectively increased the recall rate of fire image detection without compromising detection accuracy when an imbalanced dataset was used for learning. This is a major contribution of this study because recall is important, particularly in the sensing of disaster situations such as fires. In our experiments, the proposed system exhibited an accuracy of 96.9% and a recall rate of 92.9% for test image data.

• Journal of Digital Convergence
• /
• v.16 no.4
• /
• pp.1-10
• /
• 2018

In Korea, the number of recalled products is steadily increasing annually, but the recall participation rate of consumers is very low. This study looked at recall competency as a necessary factor for active recall participation by consumers. And identify the components of the recall competency and identify the recall competence factors that influence recall experience. To this end, we examined the recall experience and recall capacity of 1,626 adult consumers in Korea. As a result, five factors of recall participation will, recall related skill, recall policy recognition, subjective knowledge and objective knowledge were derived. As a result of comparing recall competencies among recall experience and non-recall experience, there were statistically significant differences in all competency factors. Recall related skill and subjective knowledge competency were significant factors for recall experience. In order to improve the effectiveness of the recall system, it is important to improve the recall information and increase access to information retrieval in order to increase the recall participation rate by strengthening the recall capacity of consumers.

• Korean Journal of Cognitive Science
• /
• v.2 no.1
• /
• pp.51-72
• /
• 1990

This study investigated the effects of macrostructure on discourse processing. In this study, three different story were employed to form macrostructures on different times, and reading time, sentence verification time and free recall rate were measured as dependent vari- ables. The results showed that forming the macrostructure influenced the reading and verification times and the recall rate. The results were interpreted to indicate that the macrostructure is important to comprehend and recall the stories, and that the story grammars, hierarchies of story schema and causal relation of sentences affect the comprehension and recall through their interaction in macrostructure.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.15 no.3
• /
• pp.197-206
• /
• 2019

Telemarketing has been used in finance due to the reduction of offline channels. In order to select telemarketing target customers, various machine learning techniques have emerged to maximize the effect of minimum cost. However, there are problems that the class imbalance, which the number of marketing success customers is smaller than the number of failed customers, and the recall rate is lower than accuracy. In this paper, we propose a method that solve the imbalanced class problem and increase the recall rate to improve the efficiency. The hybrid sampling method is applied to balance the data in the class, and the stacked deep network is applied to improve the recall and precision as well as the accuracy. The proposed method is applied to actual bank telemarketing data. As a result of the comparison experiment, the accuracy, the recall, and the precision is improved higher than that of the conventional methods.

• Nutritional Sciences
• /
• v.2 no.2
• /
• pp.113-118
• /
• 1999

Critical evaluation of energy intake data from dietary studies is difficult but important. To investigate the underreporting of total energy intake, we analyzed the one-day dietary intake data collected by 24-hour recall method from 550 elderly Koreans aged 60 years or older. Underreporting was addressed by computing the ratio of energy intake (EI) to estimated basal metabolic rate (BMRest). EI : BMRest ratio was found to be 1.38 for, men and 1.33 for women, with about 14% of men and women classified as underreporters. Underreporting of energy intake was highest in men and women who were overweight, had lower family income, or no school education. For men, the most significant variables to predict the ratio of energy intake to estimated basal metabolic. rate (EI : BMRest) were weight status, members of household, alcohol consumption and age, while income and education level were most significant for women.

• Journal of Biomedical Engineering Research
• /
• v.43 no.1
• /
• pp.45-51
• /
• 2022

The purpose of this study is to improve the classification accuracy compared to the existing InceptionV3 model by proposing a new model modified with the fully connected hierarchical structure of InceptionV3, which showed excellent performance in medical image classification. The data used for model training were trained after data augmentation on a total of 1026 chest X-ray images of patients diagnosed with normal heart and Cardiomegaly at Kyungpook National University Hospital. As a result of the experiment, the learning classification accuracy and loss of the InceptionV3 model were 99.57% and 1.42, and the accuracy and loss of the proposed model were 99.81% and 0.92. As a result of the classification performance evaluation for precision, recall, and F1 score of Inception V3, the precision of the normal heart was 78%, the recall rate was 100%, and the F1 score was 88. The classification accuracy for Cardiomegaly was 100%, the recall rate was 78%, and the F1 score was 88. On the other hand, in the case of the proposed model, the accuracy for a normal heart was 100%, the recall rate was 92%, and the F1 score was 96. The classification accuracy for Cardiomegaly was 95%, the recall rate was 100%, and the F1 score was 97. If the chest X-ray image for normal heart and Cardiomegaly can be classified using the model proposed based on the study results, better classification will be possible and the reliability of classification performance will gradually increase.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.13
• /
• pp.5143-5147
• /
• 2014

Cervical cancer screening in Malaysia is by opportunistic Pap smear which contributes to the low uptake rate. To overcome this, a pilot project called the SIPPS program (translated as information system of Pap smear program) had been introduced whereby women aged 20-65 years old are invited for Pap smear and receive recall to repeat the test. This study aimed at determining which recall method is most cost-effective in getting women to repeat Pap smear. A randomised control trial was conducted where one thousand women were recalled for repeat smear either by registered letter, phone messages, phone call or the usual postal letter. The total cost applied for cost-effectiveness analysis includes the cost of sending letter for first invitation, cost of the recall method and cost of two Pap smears. Cost-effective analysis (CEA) of Pap smear uptake by each recall method was then performed. The uptake of Pap smear by postal letter, registered letters, SMS and phone calls were 18.8%, 20.0%, 21.6% and 34.4%, respectively (p<0.05). The CER for the recall method was lowest by phone call compared to other interventions; RM 69.18 (SD RM 0.14) compared to RM 106.53 (SD RM 0.13), RM 134.02 (SD RM 0.15) and RM 136.38 (SD RM 0.11) for SMS, registered letter and letter, respectively. ICER showed that it is most cost saving if the usual method of recall by postal letter be changed to recall by phone call. The possibility of letter as a recall for repeat Pap smear to reach the women is higher compared to sending SMS or making phone call. However, getting women to do repeat Pap smear is better with phone call which allows direct communication. Despite the high cost of the phone call as a recall method for repeat Pap smear, it is the most cost-effective method compared to others.

• Korean Journal of Artificial Intelligence
• /
• v.12 no.2
• /
• pp.1-7
• /
• 2024

In this paper, we compare three models (logistic regression, Random Forest, and XGBoost) for predicting stroke occurrence using data from the Korea National Health and Nutrition Examination Survey (KNHANES). We evaluated these models using various metrics, focusing mainly on recall and F1 score to assess their performance. Initially, the logistic regression model showed a satisfactory recall score among the three models; however, it was excluded from further consideration because it did not meet the F1 score threshold, which was set at a minimum of 0.5. The F1 score is crucial as it considers both precision and recall, providing a balanced measure of a model's accuracy. Among the models that met the criteria, XGBoost showed the highest recall rate and showed excellent performance in stroke prediction. In particular, XGBoost shows strong performance not only in recall, but also in F1 score and AUC, so it should be considered the optimal algorithm for predicting stroke occurrence. This study determines that the performance of XGBoost is optimal in the field of stroke prediction.

• Journal of The Korean Association of Information Education
• /
• v.1 no.1
• /
• pp.28-37
• /
• 1997

This paper presents a new neural network based on fuzzy set and its application to invariant character recognition. The fuzzy neural network consists of five layers. The results of simulation show that the network can recognize characters in the case of distortion, translation, rotation and different sizes of handwritten characters and even with noise(8${\sim}$30%)). Translation, distortion, different sizes and noise are achieved by layer L2 and rotation invariant by layer L5. The network can recognize 108 examples of training with 100% recognition rate when they are shifted in eight directions by 1 pixel and 2 pixels. Also, the network can recognize all the distorted characters with 100% recognition rate. The simulations show that the test patterns cover a ${\pm}20^{\circ}$ range of rotation correctly. The proposed network can also recall correctly all the learned characters with 100% recognition rate. The proposed network is simple and its learning and recall speeds are very fast. This network also works for the segmentation and recognition of handwritten characters.