Just-In-Time Software Defects Prediction (JIT-SDP) plays a critical role in software engineering by enabling the early identification of potential defects before they impact system performance. This study investigates the effectiveness of Boolean Combination of Classifiers (BCC) in building effective JIT-SDP models. We propose the JITBoost framework, which leverages three BCC algorithms, namely Brute-force Boolean Combination (BBC), Iterative Boolean Combination (IBC), and Weighted Pruning Iterative Boolean Combination (WPIBC). JITBoost combines the decisions of six traditional machine learning algorithms and one deep learning algorithm. When applied to 259K commits of 34 projects, we show that JITBoost models perform better than traditional machine learning and deep learning algorithms when used individually. Specifically, JITBoost-BBC, JITBoost-IBC, and JITBoost-WPIBC achieve mean AUCs of 0.891, 0.879, and 0.886, respectively, with cross-validation. With a time-aware data-splitting approach, they achieve mean AUCs of 0.863, 0.854, and 0.857, respectively. Overall, the findings suggest that combining machine learning models within the JITBoost framework can lead to improved performance in JIT-SDP models.

Because of their contribution to the overall reliability assurance process, software logs have become important data assets for the analysis of software systems. Logs are often the only data points that can shed light on how a software system behaves once deployed. Unfortunately, logs are often unstructured data items, hindering viable analysis of their content. There are studies that aim to automatically parse large log files. The primary goal is to create templates from raw log data samples that can later be used to recognize future logs. In this paper, we propose ULP, a Unified Log Parsing tool, which is highly accurate and efficient. ULP combines string matching and local frequency analysis to parse large log files in an efficient manner. First, log events are organized into groups using a text processing method. Frequency analysis is then applied locally to instances of the same group to identify static and dynamic content of log events. When applied to 10 log datasets of the LogPai benchmark, ULP achieves an average accuracy of 89.2%, which outperforms the accuracy of four leading log parsing tools, namely Drain, Logram, SPELL and AEL. Additionally, ULP can parse up to four million log events in less than 3 minutes. ULP is available online as an open source and can be readily used by practitioners and researchers to parse effectively and efficiently large log files so as to support log analysis tasks.

Existing Just-in-Time (JIT) bug prediction techniques are designed to work on single projects. In this paper, we present ClusterCommit, a JIT bug prediction approach geared towards clusters of projects that share common libraries and functionalities. Unlike existing techniques, ClusterCommit trains a machine learning model by combining commits from a set of projects that are part of a larger cluster. Once this model is built, ClusterCommit can be used to detect buggy commits in each of these projects. When applying ClusterCommits to 16 projects that revolve around the Hadoop ecosystem and 10 projects of the Hive ecosystem, the results show that ClusterCommit achieves an F1-score of 73% and MCC of 0.44 for both clusters. These preliminary results are very promising and may lead to new JIT bug prediction techniques geared towards projects that are part of a large cluster.

Summary White Blood Cell (WBC) segmentation is one of the important topics in the medical image processing field. Many researchers proposed several clustering approaches to segment WBC from blood smear microscopic images. However, a fast and robust segmentation of WBCs is still a challenging task. In this work, we propose parallel algorithms that utilize the parallelism capabilities of the Graphics Processing Units (GPUs) to accelerate the segmentation of WBC from microscopic images. In this research, we implement the main image segmentation clustering algorithms using one thread that we run on a single CPU (sequential implementation) and using multiple threads that we run on both the CPU and the GPU (hybrid CPU‐GPU). We focus our work on the most common four segmentation algorithms: Standard K‐means (SKM), Adaptive K‐means (AKM), Fuzzy C‐means (FCM), and Fuzzy Possibilistic C‐means (FPCM). We implement these algorithms and the pre‐processing steps for WBC image segmentation in CUDA programming to take the advantages of the large number of cores in GPUs. In this work, our hybrid implementation accelerated the four studied sequential algorithms by 4X, 3.8X, 3.4X, and 3.4X, respectively, without affecting WBC segmentation quality.

Summary Classification and clustering techniques are used in different applications. Large‐scale big data applications such as social networks analysis applications need to process large data chunks in a short time. Classification and clustering tasks in such applications consume a lot of processing time. Improving the performance of classification and clustering algorithms enhances the performance of applications that use such type of algorithms. This paper introduces an approach for exploiting the graphics processing unit (GPU) platform to improve the performance of classification and clustering algorithms. The proposed approach uses two GPUs implementations, which are the pure GPU or GPU‐only implementation and the GPU‐CPU hybrid implementation. The results show that the hybrid implementation, which optimizes the subtask scheduling for both the CPU and the GPU processing elements, outperforms the approach that uses only the GPU.

Big data is a main problem for data mining methods. Fortunately, the rapid advances in affordable high performance computing platforms such as the Graphics Processing Unit (GPU) have helped researchers in reducing the execution time of many algorithms including data mining algorithms. This paper discusses the utilization of the parallelism capabilities of the GPU to improve the the performance of two common clustering algorithms, which are K-Means (KM) and Fuzzy C-Means (FCM) algorithms. Two main parallelism approaches are presented: pure and hybrid. These different versions are tested under different settings including two different GPU-equipped machines (a laptop and a server). The results show excellent improvement gains of the hybrid implementations compared with the pure parallel and sequential ones. On the laptop, the best gains of the hybrid implementations compared with the sequential ones are 11.3X for KM and 10.9X for FCM. As for the server, the best gains are 13.5X for KM and 16.3X for FCM. Moreover, the paper explores the usage of a recent memory management technique for GPU called Unified Memory (UM). The results show a decrease in the performance gain of the hybrid implementations that is equal to 44% for hybrid version of KM and 61% for FCM. On the other hand, the use of UM does introduce a small advantage for the pure parallel implementation.

Parallel programming has many benefits that can help developers and researchers to improve the performance of some algorithms to become more efficient in real life. This is especially true for systems involving medical images. Image segmentation for volume extraction is a famous segmentation process that takes long time to finish execution. In this paper, we consider a new version of the Fuzzy C-Means (FCM) segmentation algorithm (known as IT2FPCM) and provide a parallel implementation of it that is 12X time faster than the sequential implementation. The considered algorithm is based on Interval Type-2 FCM and combines fuzzy and possibilistic ideas in order to obtain higher accuracy. We conduct our experiments using two different machines and the results show that the improvement gains for both machines 11X and 12X, respectively.

The Gender Identification (GI) problem is concerned with determining the gender of the author of a given text based on its contents. The GI problem is one of the authorship profiling problems which have a wide range of applications in various fields such as marketing and security. Due to its importance, extensive research efforts have been invested in the GI problem for different languages. Unfortunately, the same cannot be said about the Arabic language despite its strategic importance and widespread. In this work, we explore the GI problem for Arabic text as a supervised learning problem. Specifically, we consider and compare two approaches for feature extraction. The first one is the Bag-Of-Words (BOW) approach while the second one is based on computing features related to sentiments and emotions. One goal of this work is to confirm the validity of the common stereotype that female authors tend to write in a more emotional way than male authors. Our results show that there is no conclusive evidence that this is true for our dataset.

Over the recent decades, bioinformatics has acquired a major concern due to the rapid growth in biological data that includes protein structures and genome sequences. Many considerable efforts have been conducted by computer scientists, mathematicians and biologists to coup with complex biological problems such as sequence alignment problem, using several techniques to formulate/model the targeted biological problems as computational problems and design algorithms to solve them in an accurate and efficient manner. Needleman-Wunsch algorithm as well as other alignment algorithms have been the subject of many studies to improve their performance due to their importance and the large scale of the data they have to handle (e.g., aligning strings of hundreds of thousands of characters). Approaches included a mixture of different parallel implementations using specialized hardware such as Graphical Processing Units (GPUs) and a vectorized approach of reading and processing the input data. In this work, a parallel implementation of NW algorithm is presented using GPU due to its efficiency and high speed, to solve the slowness problem associated with this algorithm when processing large data sets, as well as to enhance the performance of the algorithm especially when processing vectors of adjacent cells parallel to the matrix miner diagonal. The experiments show that the proposed implementation improves the performance of the algorithm by 99%.