Unnecessary Instances Research Articles

Synthetic oversampling based decision support framework to solve class imbalance problem in smoking cessation program

ABSTRACT Smoking is one of the significant avoidable risk factors for premature death. Most smokers make multiple quit attempts during their lifetime but smoking dependence is not easy and many people eventually failed quit attempts. Predicting the likelihood of success in smoking cessation program is necessary for public health. In recent years, a few numbers of decision support systems have been developed for dealing with smoking cessation based on machine learning techniques. However, the class imbalance problem is increasingly recognized as serious in real-world applications. Therefore, this paper presents a synthetic minority over-sampling technique (SMOTE) based decision support framework in order to predict the success of smoking cessation program using Korea National Health and Nutrition Examination Survey (KNHANES) dataset. We carried out experiments as follows: I) the unnecessary instances and variables have been eliminated, II) then we employed three variations of SMOTE, III) also the prediction models have been constructed. Finally, compare the prediction models to obtain the best model. Our experimental results showed that SMOTE improved the prediction performance of machine learning classifiers among evaluation metrics. Moreover, SMOTE regular based Random Forest (RF) and Naive Bayes (NB) classifiers were determined the best prediction models in real-world smoking cessation dataset. Consequently, our decision support framework can interpret the important risk factors of smoking cessation using multivariate regression analysis.

Gaussian Distribution Based Oversampling for Imbalanced Data Classification

The imbalanced data classification problem widely exists in many real-world applications. Data resampling is a promising technique to deal with imbalanced data through either oversampling or undersampling. However, the traditional data resampling approaches simply take into account the local neighbor information to generate new instances in linear ways, leading to the generation of incorrect and unnecessary instances. In this study, we propose a new data resampling technique, namely, Gaussian Distribution based Oversampling (GDO), to handle the imbalanced data for classification. In GDO, anchor instances are selected from the minority class instances in a probabilistic way by taking into account the density and distance information carried by the minority instances. Then new minority instances are generated following a Gaussian distribution model. The proposed method is validated in experimental study by comparing with seven imbalanced learning approaches on 40 data sets from the KEEL repository and 10 large data sets from the UCI repository. Experimental results show that our method outperforms the other compared methods in terms of AUC, G-mean and memory usage with an increase in running time. We also apply GDO to deal with two real imbalanced data classification problems: Internet video traffic identification and metastasis detection of esophageal cancer. The experimental results once again validate the effectiveness of our approach.

Transfer active learning by querying committee

In real applications of inductive learning for classification, labeled instances are often deficient, and labeling them by an oracle is often expensive and time-consuming. Active learning on a single task aims to select only informative unlabeled instances for querying to improve the classification accuracy while decreasing the querying cost. However, an inevitable problem in active learning is that the informative measures for selecting queries are commonly based on the initial hypotheses sampled from only a few labeled instances. In such a circumstance, the initial hypotheses are not reliable and may deviate from the true distribution underlying the target task. Consequently, the informative measures will possibly select irrelevant instances. A promising way to compensate this problem is to borrow useful knowledge from other sources with abundant labeled information, which is called transfer learning. However, a significant challenge in transfer learning is how to measure the similarity between the source and the target tasks. One needs to be aware of different distributions or label assignments from unrelated source tasks; otherwise, they will lead to degenerated performance while transferring. Also, how to design an effective strategy to avoid selecting irrelevant samples to query is still an open question. To tackle these issues, we propose a hybrid algorithm for active learning with the help of transfer learning by adopting a divergence measure to alleviate the negative transfer caused by distribution differences. To avoid querying irrelevant instances, we also present an adaptive strategy which could eliminate unnecessary instances in the input space and models in the model space. Extensive experiments on both the synthetic and the real data sets show that the proposed algorithm is able to query fewer instances with a higher accuracy and that it converges faster than the state-of-the-art methods.