• Journal of Integrative Natural Science
• /
• v.11 no.3
• /
• pp.161-164
• /
• 2018

Over the past decade, mass spectrometry-based metabolomics, especially two dimensional gas chromatography mass spectrometry (GCxGC/TOF-MS), has become a key analytical tool for metabolomics data because of its sensitivity and ability to analyze complex biological or biochemical sample. However, the need to reduce variations within/between experiments has been reported and methodological developments to overcome such problem has long been a critical issue. Along with methodological developments, developing reasonable performance measure has also been studied. Following four numerical measures have been typically used for comparison: sensitivity, specificity, receiver operating characteristic (ROC) curves, and positive predictive value (PPV). However, more recently, such measures are replaced with F1 score in many fields including metabolomics area without any carefulness of its validity. Thus, we want to investigate the validity of F1 score on two examples, with the goal of raising the awareness in choosing appropriate performance comparison measure. We noticed that F1 score itself, as a performance measure, was not good enough. Accordingly, we suggest that F1 score be supplemented with other performance measure such as specificity to improve its validity.

• Asian-Australasian Journal of Animal Sciences
• /
• v.19 no.2
• /
• pp.266-272
• /
• 2006

Numerous reports have evaluated the predictive ability of carcass probes for meat quality using measurements taken early postmortem or near 24 h. The intervening time period, however, has been largely ignored. In this study, the capacity of three probes [pH, electrical conductivity (EC), and grading probe light reflectance (GP)] to predict pork longissimus muscle quality (drip and cooking losses, Warner-Bratzler shear, $L^*$, n = 30) was evaluated at 45 min, 90 min, 3, 6, 12, 24, and 48 h postmortem. The strongest relationships were observed between cooking loss and 6 h EC and GP ($R^2$ = 0.66, 0.72), and $L^*$ and GP ($R^2$ = 0.57-0.66, 12-48 h). pH was most valuable early postmortem ($R^2$ = 0.63, 90 min with cooking loss). GP at 6 h most effectively ($R^2$ = 0.84) predicted a two factor (cooking loss+$L^*$) meat quality index. Results emphasize the predictive value of measures taken between 3 and 12 h postmortem.

• Asia pacific journal of information systems
• /
• v.25 no.1
• /
• pp.83-97
• /
• 2015

Recently, the exhibition industry has developed rapidly with the development of information technologies. Most exhibitors in an exhibition plan and deploy many events that may provide advantages to visitors as a method of effective promotion. The growth and propagation of wireless technologies is a powerful marketing tool for exhibitors. However, exhibitors still rely on domain experts who are costly and time consuming because of the manual knowledge input procedure. Moreover, it is prone to biases and errors and not suitable for managing fast-growing and tremendous amounts of data that far exceed a human's ability to comprehend. To overcome these problems, data mining technology may be a great alternative, but it needs to be fit to each exhibition. This study uses data mining technology with the Predictive Model Markup Language (PMML) to suggest a system that supports intelligent services and that improves stakeholder satisfaction. This system provides advantages to the exhibitor, show organizer, and system designer, and is first enhanced by integrating data mining technologies through the knowledge of exhibition experts. Second, using the PMML, the system can automate the process of applying data mining models to solve real-time processing problems in the exhibition environment.

• Experimental and Molecular Medicine
• /
• v.50 no.12
• /
• pp.10.1-10.11
• /
• 2018

Cancer growth and progression are associated with immune suppression. Cancer cells have the ability to activate different immune checkpoint pathways that harbor immunosuppressive functions. Monoclonal antibodies that target immune checkpoints provided an immense breakthrough in cancer therapeutics. Among the immune checkpoint inhibitors, PD-1/PD-L1 and CTLA-4 inhibitors showed promising therapeutic outcomes, and some have been approved for certain cancer treatments, while others are under clinical trials. Recent reports have shown that patients with various malignancies benefit from immune checkpoint inhibitor treatment. However, mainstream initiation of immune checkpoint therapy to treat cancers is obstructed by the low response rate and immune-related adverse events in some cancer patients. This has given rise to the need for developing sets of biomarkers that predict the response to immune checkpoint blockade and immune-related adverse events. In this review, we discuss different predictive biomarkers for anti-PD-1/PD-L1 and anti-CTLA-4 inhibitors, including immune cells, PD-L1 overexpression, neoantigens, and genetic and epigenetic signatures. Potential approaches for further developing highly reliable predictive biomarkers should facilitate patient selection for and decision-making related to immune checkpoint inhibitor-based therapies.

• Biomedical Science Letters
• /
• v.25 no.3
• /
• pp.256-266
• /
• 2019

In this study, obese adults were compared for their ability to predict obesity and lipid related variables and their optimal cutoff values to predict metabolic syndrome and insulin resistance. In this study, 9,256 adults aged 20 years or older and less than 80 years old, who were in the Gyeonggi region from January 2014 to December 2016 and who were examined at a general hospital, were enrolled. The diagnostic criteria for obesity were WHO (World Health Organization), and BMI $25kg/m^2$ or more presented in the Asia-Pacific region. Metabolic syndrome was diagnosed based on the criteria of American Heart Association / National Heart, Lung, and Blood Institute (AHA / NHLBI). According to the results of receiver operating characteristic curve (ROC) analysis, Triglyceride / HDL-cholesterol (TG / HDL-C), Triglyceride and Glucose (TyG) index, lipid accumulation product (LAP) and visceral adiposity index (VAI) showed high predictive power for diagnosing metabolic syndrome. The diagnostic accuracy of LAP (AUC: 0.854) for males and VAI (0.888) for females was the highest. The optimal cutoff value of LAP was 42.71 for male and 35.44 for female, and the cutoff value of VAI was 1.92 for male and 2.15 for female. In addition, WHtR (waist to height ratio), TyG index, and LAP were used as predictors of insulin resistance in obese adults. Therefore, LAP and VAI were superior to other indicators in predicting metabolic syndrome in obese adults.

• Safety and Health at Work
• /
• v.12 no.1
• /
• pp.1-9
• /
• 2021

This review article addresses the role of safety professionals in the diffusion strategies for predictive analytics for safety performance. The article explores the models, definitions, roles, and relationships of safety professionals in knowledge application, access, management, and leadership in safety analytics. The article addresses challenges safety professionals face when integrating safety analytics in organizational settings in four operations areas: application, technology, management, and strategy. A review of existing conventional safety data sources (safety data, internal data, external data, and context data) is briefly summarized as a baseline. For each of these data sources, the article points out how emerging analytic data sources (such as Industry 4.0 and the Internet of Things) broaden and challenge the scope of work and operational roles throughout an organization. In doing so, the article defines four perspectives on the integration of predictive analytics into organizational safety practice: the programmatic perspective, the technological perspective, the sociocultural perspective, and knowledge-organization perspective. The article posits a four-level, organizational knowledge-skills-abilities matrix for analytics integration, indicating key organizational capacities needed for each area. The work shows the benefits of organizational alignment, clear stakeholder categorization, and the ability to predict future safety performance.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.3
• /
• pp.145-154
• /
• 2024

In order to operate predictive models utilizing administrative big data, it is crucial to consider policy changes and the characteristics of highly volatile data. Considering this scenario, this study proposes the Cut-off Voting Classifier (CVC) algorithm. This proposed algorithm prevents a sharp decline in accuracy by utilizing multiple weak classifiers. The study validates the proposed algorithm's performance through experiments. The performance evaluation demonstrates the ability to maintain stable prediction rates even in situations with a sharp decline in predictive model accuracy.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.4
• /
• pp.526-533
• /
• 2014

In this study, we introduce design methodology to develop a guidance for issuing heavy rainfall warning by using both RBFNNs(Radial basis function neural networks) and SVR(Support vector regression) model, and then carry out the comparative studies between two pattern classifiers. Individual classifiers are designed as architecture realized with the aid of optimization and pre-processing algorithm. Because the predictive performance of the existing heavy rainfall forecast system is commonly affected from diverse processing techniques of meteorological data, under-sampling method as the pre-processing method of input data is used, and also data discretization and feature extraction method for SVR and FCM clustering and PSO method for RBFNNs are exploited respectively. The observed data, AWS(Automatic weather wtation), supplied from KMA(korea meteorological administration), is used for training and testing of the proposed classifiers. The proposed classifiers offer the related information to issue a heavy rain warning in advance before 1 to 3 hours by using the selected meteorological data and the cumulated precipitation amount accumulated for 1 to 12 hours from AWS data. For performance evaluation of each classifier, ETS(Equitable Threat Score) method is used as standard verification method for predictive ability. Through the comparative studies of two classifiers, neuro-fuzzy method is effectively used for improved performance and to show stable predictive result of guidance to issue heavy rainfall warning.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.3
• /
• pp.62-67
• /
• 1999

In this paper, we investigates a reliable model of the Predictive Recurrent Neural Network for the speech recognition. Predictive Neural Networks are modeled by syllable units. For the given input syllable, then a model which gives the minimum prediction error is taken as the recognition result. The Predictive Neural Network which has the structure of recurrent network was composed to give the dynamic feature of the speech pattern into the network. We have compared with the recognition ability of the Recurrent Network proposed by Elman and Jordan. ETRI's SAMDORI has been used for the speech DB. In order to find a reliable model of neural networks, the changes of two recognition rates were compared one another in conditions of: (1) changing prediction order and the number of hidden units: and (2) accumulating previous values with self-loop coefficient in its context. The result shows that the optimum prediction order, the number of hidden units, and self-loop coefficient have differently responded according to the structure of neural network used. However, in general, the Jordan's recurrent network shows relatively higher recognition rate than Elman's. The effects of recognition rate on the self-loop coefficient were variable according to the structures of neural network and their values.

• Journal of Internet Computing and Services
• /
• v.12 no.6
• /
• pp.83-94
• /
• 2011

Main researching issue in affective computing is to give a machine the ability to recognize the emotion of a person and to react it properly. Efforts in that direction have mainly focused on facial and oral cues to get emotions. Postures have been recently considered as well. This paper aims to discriminate emotions posture by identifying and measuring the saliency of posture features that play a role in affective expression. To do so, affective postures from human subjects are first collected using a motion capture system, then emotional features in posture are described with spatial ones. Through standard statistical techniques, we verified that there is a statistically significant correlation between the emotion intended by the acting subjects, and the emotion perceived by the observers. Discriminant Analysis are used to build affective posture predictive models and to measure the saliency of the proposed set of posture features in discriminating between 6 basic emotional states. The evaluation of proposed features and models are performed using a correlation between actor-observer's postures set. Quantitative experimental results show that proposed set of features discriminates well between emotions, and also that built predictive models perform well.