Detection of fraudulent credit card transactions: A comparative analysis of data sampling and classification techniques

Every year there is an increasing loss of a huge amount of money due to fraudulent credit card transactions. Recently there is a focus on using machine learning algorithms to identify fraud transactions. The number of fraud cases to non-fraud transactions is very low. This creates a skewed or unbalanced data, which poses a challenge to training the machine learning models. The availability of a public dataset for this research problem is scarce. The dataset used for this work is obtained from Kaggle. In this paper, we explore different sampling techniques such as under-sampling, Synthetic Minority Oversampling Technique (SMOTE) and SMOTE-Tomek, to work on the unbalanced data. Classification models, such as k-Nearest Neighbour (KNN), logistic regression, random forest and Support Vector Machine (SVM), are trained on the sampled data to detect fraudulent credit card transactions. The performance of the various machine learning approaches are evaluated for its precision, recall and F1-score. The classification results obtained is promising and can be used for credit card fraud detection.

Garbage Classification Using Ensemble DenseNet169

Garbage is a big problem for the sustainability of the environment, economy, and society, where the demand for waste increases along with the growth of society and its needs. Where in 2019 Indonesia was able to produce 66-67 million tons of waste, which is an increase from the previous year of 2 to 3 million tons of waste. Waste management efforts have been carried out by the government, including by making waste sorting regulations. This sorting is known as 3R (reduce, reuse, recycle), but most people do not sort their waste properly. In this study, a model was developed that can sort out 6 types of waste including: cardboard, glass, metal, paper, plastic, trash. The model was built using the transfer learning method with a pretrained model DenseNet169. Where the optimal results are shown for the classes that have been oversampling previously with an accuracy of 91%, an increase of 1% compared to the model that has an unbalanced data distribution. The next model optimization is done by applying the ensemble method to the four models that have been oversampled on the training dataset with the same architecture. This method shows an increase of 3% to 5% while the final accuracy on the test of dataset is 96%.

Methodology for eliminating imbalance of image data sets

The methodology of solving the problem of eliminating class imbalance in image data sets is presented. The proposed methodology includes the stages of image fragment extraction, fragment augmentation, feature extraction, duplication of minority objects, and is based on reinforcement learning technology. The degree of imbalance indicator was used as a measure to determine the imbalance of the data set. An experiment was performed using a set of images of the faces of patients with skin rashes, annotated according to the severity of acne. The main steps of the methodology implementation are considered. The results of the classification showed the feasibility of applying the proposed methodology. The accuracy of classification on test data was 85%, which is 5% higher than the result obtained without the use of the proposed methodology. Key words: class imbalance, unbalanced data set, image fragment extraction, augmentation.

Fine-Grained Named Entity Recognition Using a Multi-Stacked Feature Fusion and Dual-Stacked Output in Korean

Named entity recognition (NER) is a natural language processing task to identify spans that mention named entities and to annotate them with predefined named entity classes. Although many NER models based on machine learning have been proposed, their performance in terms of processing fine-grained NER tasks was less than acceptable. This is because the training data of a fine-grained NER task is much more unbalanced than those of a coarse-grained NER task. To overcome the problem presented by unbalanced data, we propose a fine-grained NER model that compensates for the sparseness of fine-grained NEs by using the contextual information of coarse-grained NEs. From another viewpoint, many NER models have used different levels of features, such as part-of-speech tags and gazetteer look-up results, in a nonhierarchical manner. Unfortunately, these models experience the feature interference problem. Our solution to this problem is to adopt a multi-stacked feature fusion scheme, which accepts different levels of features as its input. The proposed model is based on multi-stacked long short-term memories (LSTMs) with a multi-stacked feature fusion layer for acquiring multilevel embeddings and a dual-stacked output layer for predicting fine-grained NEs based on the categorical information of coarse-grained NEs. Our experiments indicate that the proposed model is capable of state-of-the-art performance. The results show that the proposed model can effectively alleviate the unbalanced data problem that frequently occurs in a fine-grained NER task. In addition, the multi-stacked feature fusion layer contributes to the improvement of NER performance, confirming that the proposed model can alleviate the feature interference problem. Based on this experimental result, we conclude that the proposed model is well-designed to effectively perform NER tasks.

Genetic Programming for Classification with Unbalanced Data

<p>In classification,machine learning algorithms can suffer a performance bias when data sets are unbalanced. Binary data sets are unbalanced when one class is represented by only a small number of training examples (called the minority class), while the other class makes up the rest (majority class). In this scenario, the induced classifiers typically have high accuracy on the majority class but poor accuracy on the minority class. As the minority class typically represents the main class-of-interest in many real-world problems, accurately classifying examples from this class can be at least as important as, and in some cases more important than, accurately classifying examples from the majority class. Genetic Programming (GP) is a promising machine learning technique based on the principles of Darwinian evolution to automatically evolve computer programs to solve problems. While GP has shown much success in evolving reliable and accurate classifiers for typical classification tasks with balanced data, GP, like many other learning algorithms, can evolve biased classifiers when data is unbalanced. This is because traditional training criteria such as the overall success rate in the fitness function in GP, can be influenced by the larger number of examples from the majority class. This thesis proposes a GP approach to classification with unbalanced data. The goal is to develop new internal cost-adjustment techniques in GP to improve classification performances on both the minority class and the majority class. By focusing on internal cost-adjustment within GP rather than the traditional databalancing techniques, the unbalanced data can be used directly or "as is" in the learning process. This removes any dependence on a sampling algorithm to first artificially re-balance the input data prior to the learning process. This thesis shows that by developing a number of new methods in GP, genetic program classifiers with good classification ability on the minority and the majority classes can be evolved. This thesis evaluates these methods on a range of binary benchmark classification tasks with unbalanced data. This thesis demonstrates that unlike tasks with multiple balanced classes where some dynamic (non-static) classification strategies perform significantly better than the simple static classification strategy, either a static or dynamic strategy shows no significant difference in the performance of evolved GP classifiers on these binary tasks. For this reason, the rest of the thesis uses this static classification strategy. This thesis proposes several new fitness functions in GP to perform cost adjustment between the minority and the majority classes, allowing the unbalanced data sets to be used directly in the learning process without sampling. Using the Area under the Receiver Operating Characteristics (ROC) curve (also known as the AUC) to measure how well a classifier performs on the minority and majority classes, these new fitness functions find genetic program classifiers with high AUC on the tasks on both classes, and with fast GP training times. These GP methods outperform two popular learning algorithms, namely, Naive Bayes and Support Vector Machines on the tasks, particularly when the level of class imbalance is large, where both algorithms show biased classification performances. This thesis also proposes a multi-objective GP (MOGP) approach which treats the accuracies of the minority and majority classes separately in the learning process. The MOGP approach evolves a good set of trade-off solutions (a Pareto front) in a single run that perform as well as, and in some cases better than, multiple runs of canonical single-objective GP (SGP). In SGP, individual genetic program solutions capture the performance trade-off between the two objectives (minority and majority class accuracy) using an ROC curve; whereas in MOGP, this requirement is delegated to multiple genetic program solutions along the Pareto front. This thesis also shows how multiple Pareto front classifiers can be combined into an ensemble where individual members vote on the class label. Two ensemble diversity measures are developed in the fitness functions which treat the diversity on both the minority and the majority classes as equally important; otherwise, these measures risk being biased toward the majority class. The evolved ensembles outperform their individual members on the tasks due to good cooperation between members. This thesis further improves the ensemble performances by developing a GP approach to ensemble selection, to quickly find small groups of individuals that cooperate very well together in the ensemble. The pruned ensembles use much fewer individuals to achieve performances that are as good as larger (unpruned) ensembles, particularly on tasks with high levels of class imbalance, thereby reducing the total time to evaluate the ensemble.</p>

Genetic Programming for Classification with Unbalanced Data

<p>In classification,machine learning algorithms can suffer a performance bias when data sets are unbalanced. Binary data sets are unbalanced when one class is represented by only a small number of training examples (called the minority class), while the other class makes up the rest (majority class). In this scenario, the induced classifiers typically have high accuracy on the majority class but poor accuracy on the minority class. As the minority class typically represents the main class-of-interest in many real-world problems, accurately classifying examples from this class can be at least as important as, and in some cases more important than, accurately classifying examples from the majority class. Genetic Programming (GP) is a promising machine learning technique based on the principles of Darwinian evolution to automatically evolve computer programs to solve problems. While GP has shown much success in evolving reliable and accurate classifiers for typical classification tasks with balanced data, GP, like many other learning algorithms, can evolve biased classifiers when data is unbalanced. This is because traditional training criteria such as the overall success rate in the fitness function in GP, can be influenced by the larger number of examples from the majority class. This thesis proposes a GP approach to classification with unbalanced data. The goal is to develop new internal cost-adjustment techniques in GP to improve classification performances on both the minority class and the majority class. By focusing on internal cost-adjustment within GP rather than the traditional databalancing techniques, the unbalanced data can be used directly or "as is" in the learning process. This removes any dependence on a sampling algorithm to first artificially re-balance the input data prior to the learning process. This thesis shows that by developing a number of new methods in GP, genetic program classifiers with good classification ability on the minority and the majority classes can be evolved. This thesis evaluates these methods on a range of binary benchmark classification tasks with unbalanced data. This thesis demonstrates that unlike tasks with multiple balanced classes where some dynamic (non-static) classification strategies perform significantly better than the simple static classification strategy, either a static or dynamic strategy shows no significant difference in the performance of evolved GP classifiers on these binary tasks. For this reason, the rest of the thesis uses this static classification strategy. This thesis proposes several new fitness functions in GP to perform cost adjustment between the minority and the majority classes, allowing the unbalanced data sets to be used directly in the learning process without sampling. Using the Area under the Receiver Operating Characteristics (ROC) curve (also known as the AUC) to measure how well a classifier performs on the minority and majority classes, these new fitness functions find genetic program classifiers with high AUC on the tasks on both classes, and with fast GP training times. These GP methods outperform two popular learning algorithms, namely, Naive Bayes and Support Vector Machines on the tasks, particularly when the level of class imbalance is large, where both algorithms show biased classification performances. This thesis also proposes a multi-objective GP (MOGP) approach which treats the accuracies of the minority and majority classes separately in the learning process. The MOGP approach evolves a good set of trade-off solutions (a Pareto front) in a single run that perform as well as, and in some cases better than, multiple runs of canonical single-objective GP (SGP). In SGP, individual genetic program solutions capture the performance trade-off between the two objectives (minority and majority class accuracy) using an ROC curve; whereas in MOGP, this requirement is delegated to multiple genetic program solutions along the Pareto front. This thesis also shows how multiple Pareto front classifiers can be combined into an ensemble where individual members vote on the class label. Two ensemble diversity measures are developed in the fitness functions which treat the diversity on both the minority and the majority classes as equally important; otherwise, these measures risk being biased toward the majority class. The evolved ensembles outperform their individual members on the tasks due to good cooperation between members. This thesis further improves the ensemble performances by developing a GP approach to ensemble selection, to quickly find small groups of individuals that cooperate very well together in the ensemble. The pruned ensembles use much fewer individuals to achieve performances that are as good as larger (unpruned) ensembles, particularly on tasks with high levels of class imbalance, thereby reducing the total time to evaluate the ensemble.</p>

Genetic Programming for Binary Classification with High-dimensional Unbalanced Data

<p><b>Class imbalance and high dimensionality have been acknowledged as two tough issues in classification. Learning from unbalanced data, the constructed classifiers are often biased towards the majority class, and thereby perform poorly on the minority class. Unfortunately, the minority class is often the class of interest in many real-world applications, such as medical diagnosis and fault detection. High dimensionality often makes it more difficult to handle the class imbalance issue. To date, most existing works attempt to address one single issue, without consideration of solving the other. These works could not be effectively applied to some challenging classification tasks that suffer from both of the two issues.</b></p> <p>Genetic programming (GP) is one of the most popular techniques from evolutionary computation, which has been widely applied to classification tasks. The built-in feature selection ability of GP makes it very powerful for use in classification with high-dimensional data. However, if the class imbalance issue is not well addressed, the constructed GP classifiers are often biased towards the majority class. Accordingly, this thesis aims to address the joint effects of class imbalance and high dimensionality by developing new GP based classification approaches, with the goal of improving classification performance.</p> <p>To effectively and efficiently address the performance bias issue of GP, this thesis develops a fitness function that considers two criteria, namely the approximation of area under the curve (AUC) and classification clarity (i.e. how well a program can separate the two classes). To further improve the efficiency, a new program reuse mechanism is designed to reuse previous effective GP individuals. According to experimental results, GP with the new fitness function and the program reuse mechanism achieves good performance and significantly saves training time. However, this method treats the two criteria equally, which is not always reasonable.</p> <p>To avoid manually weighing the two criteria in the fitness evaluation process, we propose a novel two-criterion fitness evaluation method, where the obtained values on the two criteria are combined in pairs, instead of summing them together. Then, a three-criterion tournament selection is designed to effectively identify and select good programs to be used by genetic operators for generating better offspring during the evolutionary learning process. Experimental results show that the proposed GP method achieves better classification performance than compared methods.</p> <p>Cost-sensitive learning is a popular approach to addressing the problem of class imbalance for many classification algorithms in machine learning. However, cost-sensitive algorithms are dependent on cost matrices that are usually designed manually. Unfortunately, it is often not easy for humans, even experts, to accurately specify misclassification costs for different mistakes due to the lack or incompleteness of domain knowledge related to actual situations in many complex tasks. As a result, these cost-sensitive algorithms cannot be directly applied. This thesis develops new GP based approaches to developing cost-sensitive classifiers without requiring cost matrices from humans. The newly developed cost-sensitive GP methods are able to construct classifiers and learn cost values or intervals automatically and simultaneously. The experimental results show that the new cost-sensitive GP methods outperform compared methods for high-dimensional unbalanced classification in almost all comparisons.</p> <p>Cost-sensitive GP classifiers treat the minority class as being more important than the majority class, but this may cause an accuracy decrease in the overlapping areas where the prior probabilities of the two classes are about the same. In the thesis, we propose a neighborhood method to detect overlapping areas, and then use GP to develop cost-sensitive classifiers that employ different classification strategies to classify instances from the overlapping areas or the non-overlapping areas.</p>