Predict-then-optimize or predict-and-optimize? An empirical evaluation of cost-sensitive learning strategies

Vanderschueren, Toon; Verdonck, Tim; Baesens, Bart; Verbeke, Wouter

doi:10.1016/j.ins.2022.02.021

Cited by 32 publications

(22 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Cost-sensitive learning reduces the error of SVM in classifying minority classes by reducing the overall cost of misclassification and improves the classification of unbalanced data [ 26 ]. Vanderschueren T [ 27 ] improved the general learning model into the cost-sensitive learning model by calculating the ideal cost for each sample and modifying the original sample class to obtain the new sample set. Ren Z [ 28 ] used fuzzy learning to reduce the effect of noise in samples on classification and combined the cost-sensitive mechanism to reduce the sensitivity of unbalance distribution.…”

Section: Introductionmentioning

confidence: 99%

Prediction and optimization of employee turnover intentions in enterprises based on unbalanced data

Li,

Fox

2023

PLoS ONE

View full text Add to dashboard Cite

The sudden resignation of core employees often brings losses to companies in various aspects. Traditional employee turnover theory cannot analyze the unbalanced data of employees comprehensively, which leads the company to make wrong decisions. In the face the classification of unbalanced data, the traditional Support Vector Machine (SVM) suffers from insufficient decision plane offset and unbalanced support vector distribution, for which the Synthetic Minority Oversampling Technique (SMOTE) is introduced to improve the balance of generated data. Further, the Fuzzy C-mean (FCM) clustering is improved and combined with the SMOTE (IFCM-SMOTE-SVM) to new synthesized samples with higher accuracy, solving the drawback that the separation data synthesized by SMOTE is too random and easy to generate noisy data. The kernel function is combined with IFCM-SMOTE-SVM and transformed to a high-dimensional space for clustering sampling and classification, and the kernel space-based classification algorithm (KS-IFCM-SMOTE-SVM) is proposed, which improves the effectiveness of the generated data on SVM classification results. Finally, the generalization ability of KS-IFCM-SMOTE-SVM for different types of enterprise data is experimentally demonstrated, and it is verified that the proposed algorithm has stable and accurate performance. This study introduces the SMOTE and FCM clustering, and improves the SVM by combining the data transformation in the kernel space to achieve accurate classification of unbalanced data of employees, which helps enterprises to predict whether employees have the tendency to leave in advance.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction and optimization of employee turnover intentions in enterprises based on unbalanced data

Li,

Fox

2023

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…Thresholding is a postprocessing technique employed in CSL to fine-tune the classification decisions of a trained model based on specific cost considerations. Unlike other approaches, thresholding operates on the output probability estimates produced by a model after training during the test phase (Fernández et al 2018;Johnson and Khoshgoftaar 2019;Vanderschueren et al 2022). It serves as a meta-learning approach, allowing the conversion of any cost-insensitive model into a cost-sensitive one (Johnson and Khoshgoftaar 2019;Sheng and Ling 2006), and possesses the advantage of being more accessible and comprehensible to practitioners (Feng et al 2020).…”

Section: Thresholdingmentioning

confidence: 99%

“…In addition to the technique described above, another method for threshold optimisation is empirical thresholding (Zhao et al 2018;Reychav et al 2019). This method involves iteratively searching for the optimal threshold that minimises the total cost on a validation set (Vanderschueren et al 2022), offering an alternative means of optimising cost-sensitive classification.…”

Section: Thresholdingmentioning

confidence: 99%

“…Expanding upon our exploration of the strengths and weaknesses of CSL approaches, it is crucial to consider the valuable insights derived from other studies in the field. Recent research (Vanderschueren et al 2022) categorises CSL approaches based on the stage at which misclassification costs are incorporated into two classes: cost-sensitive training of models and cost-sensitive decision-making. The former category encompasses techniques applied before or during model training to construct a classifier, notably direct approaches and meta-learning preprocessing methods.…”

Section: Strengths and Weaknesses Of Csl Approachesmentioning

confidence: 99%

See 1 more Smart Citation

Cost-sensitive learning for imbalanced medical data: a review

Araf,

Idri,

Chairi

2024

Artif Intell Rev

View full text Add to dashboard Cite

Integrating Machine Learning (ML) in medicine has unlocked many opportunities to harness complex medical data, enhancing patient outcomes and advancing the field. However, the inherent imbalanced distribution of medical data poses a significant challenge, resulting in biased ML models that perform poorly on minority classes. Mitigating the impact of class imbalance has prompted researchers to explore various strategies, wherein Cost-Sensitive Learning (CSL) arises as a promising approach to improve the accuracy and reliability of ML models. This paper presents the first review of CSL for imbalanced medical data. A comprehensive exploration of the existing literature encompassed papers published from January 2010 to December 2022 and sourced from five major digital libraries. A total of 173 papers were selected, analysed, and classified based on key criteria, including publication years, channels and sources, research types, empirical types, medical sub-fields, medical tasks, CSL approaches, strengths and weaknesses of CSL, frequently used datasets and data types, evaluation metrics, and development tools. The results indicate a noteworthy publication rise, particularly since 2020, and a strong preference for CSL direct approaches. Data type analysis unveiled diverse modalities, with medical images prevailing. The underutilisation of cost-related metrics and the prevalence of Python as the primary programming tool are highlighted. The strengths and weaknesses analysis covered three aspects: CSL strategy, CSL approaches, and relevant works. This study serves as a valuable resource for researchers seeking to explore the current state of research, identify strengths and gaps in the existing literature and advance CSL’s application for imbalanced medical data.

show abstract

“…To address uncertainty, the problem is typically tackled using a two-stage method. In the first stage, demand prediction is performed, and in the second stage, the predicted value is used in an optimization algorithm to make decisions [4]. But enhancing prediction accuracy doesn't always guarantee optimal decision-making [5].…”

Section: Introductionmentioning

confidence: 99%

Readiness Assessment of Public University in Bangladesh for Education 4.0: A Case Study of Shahjalal University of Science and Technology – Sylhet, Bangladesh

Islam

2022

Khulna Univ. Stud.

View full text Add to dashboard Cite

In order to get the benefits of Industry 4.0 (4ir) we need the graduates who are able to face the challenges of 4ir. As the universities are the main sources of producing graduates, they should produce adaptable graduates of 4ir. The 4ir demands such a system of higher education where each and every stakeholders especially educators and learners should be connected in a globalized automated environment, networked, virtualized and flexible. In the paradigm of 4ir learners should learn to learn the sources of learning how to build knowledge and skills within the process of teaching –learning. The study aims to identify the dimensions of E4.0 corresponding to 4ir and to assess the readiness of Shahjalal University of Science and Technology (SUST) is one of 50 public universities of Bangladesh. At present there are 28 disciplines in six schools of studies in SUST and among them 20 departments have been selected for the study based on minimum 10 years of establishment. The assessment has been focused on four aspects mainly curriculum, teacher (faculty), intuitional qualification and logistic supports in terms of E4.0. In conducting the study data from collected 120 teachers and graduate students with a structured questionnaire. The questionnaire was five point scales. The collected data was analyzed with SPSS and mainly frequency in each dimension was found and finally factor analysis was done to find the factors affecting E4.0. The frequency distribution showed that Robotic Process Automation aspects ranked first following cloud computing, data analytics, IoT, AI, and BCT. The factor analysis showed that seventeen aspects have been considered as important and the top most five aspects are fees and rewards, data as a service, software as a service, AI extraction and user device.

show abstract

Predict-then-optimize or predict-and-optimize? An empirical evaluation of cost-sensitive learning strategies

Cited by 32 publications

References 42 publications

Prediction and optimization of employee turnover intentions in enterprises based on unbalanced data

Prediction and optimization of employee turnover intentions in enterprises based on unbalanced data

Cost-sensitive learning for imbalanced medical data: a review

Readiness Assessment of Public University in Bangladesh for Education 4.0: A Case Study of Shahjalal University of Science and Technology – Sylhet, Bangladesh

Contact Info

Product

Resources

About