Electronic transaction fraud has been a severe threat in recent years, causing substantial financial losses and devaluing the reputation of financial institutions. Various machinelearning and deep learning models hav...
详细信息
Due to the increasing power of computers and the algorithmic advances that have occurred in the field of machinelearning, techniques for analyzing data have become more powerful tools. Quantum computers could perform...
详细信息
Because of the naturalness of software and the rapid evolution of machinelearning (ML) techniques, frequently repeated code change patterns (CPATs) occur often. They range from simple API migrations to changes involv...
详细信息
ISBN:
(纸本)9781665457019
Because of the naturalness of software and the rapid evolution of machinelearning (ML) techniques, frequently repeated code change patterns (CPATs) occur often. They range from simple API migrations to changes involving several complex control structures such as for loops. While manually performing CPATs is tedious, the current state-of-the-art techniques for inferring transformation rules are not advanced enough to handle unseen variants of complex CPATs, resulting in a low recall rate. In this paper we present a novel, automated workflow that mines CPATs, infers the transformation rules, and then transplants them automatically to new target sites. We designed, implemented, evaluated and released this in a tool, PYEVOLVE. At its core is a novel data-flow, control-flow aware transformation rule inference engine. Our technique allows us to advance the state-of-the-art for transformation-by-example tools;without it, 70% of the code changes that PYEVOLVE transforms would not be possible to automate. Our thorough empirical evaluation of over 40,000 transformations shows 97% precision and 94% recall. By accepting 90% of CPATs generated by PYEVOLVE in famous open-source projects, developers confirmed its changes are useful.
software cost estimation is a challenging and complex task during software development. It directs project managers and developers to analyze and predict costs at the beginning of the software development life cycle. ...
详细信息
The proceedings contain 20 papers. The topics discussed include: softwareengineering as the linchpin of responsible ai;one adapter for all programming languages? adapter tuning for code search and summarization;keepi...
ISBN:
(纸本)9798350322613
The proceedings contain 20 papers. The topics discussed include: softwareengineering as the linchpin of responsible ai;one adapter for all programming languages? adapter tuning for code search and summarization;keeping pace with ever-increasing data: towards continual learning of code intelligence models;detecting JVM JIT compiler bugs via exploring two-dimensional input spaces;validating SMT solvers via skeleton enumeration empowered by historical bug-triggering inputs;regression fuzzing for deep learning systems;the untold story of code refactoring customizations in practice;a comprehensive study of real-world bugs in machinelearning model optimization;evaluating the impact of experimental assumptions in automated fault localization;locating framework-specific crashing faults with compact and explainable candidate set;and how do we read formal claims? eye-tracking and the cognition of proofs about algorithms.
In this project, we are evaluating the efficacy of utilizing deep learning and convolutional neural networks to support individuals who have visual disabilities. Our objective is to create a mobile or web application ...
详细信息
Issue tracking systems comprise data which are useful in evaluating or improving software development processes. Revealing and interpreting this information is a challenging problem which needs appropriate algorithms ...
详细信息
ISBN:
(纸本)9789897586477
Issue tracking systems comprise data which are useful in evaluating or improving software development processes. Revealing and interpreting this information is a challenging problem which needs appropriate algorithms and tools. For this purpose, we use text mining schemes adapted to the specificity of the software repository. They base on a detailed analysis of the used dictionaries which comprise Natural Language Words (NLW) and are enhanced with specialized entities in issue descriptions (e.g., emails, code snippets, technical names). They are defined with specially developed regular expressions. The pre-processed texts are submitted to original text mining algorithms (machinelearning). This approach has been verified in commercial and open-source projects and showed possible development improvements.
This study aims to propose an FPGA-oriented ECG signal processing system architecture. The method identifies FPGA complexes and classifies beats as either preterm or typical contractions of the ventricular wall. For p...
详细信息
By studying the basic principle of ICP and the characteristics of low-temperature plasma, the advantages of using a plasma metal cylinder discharge chamber are discussed, and the influence of skin effect on discharge ...
详细信息
Although Computer Numerical Control (CNC) machines were designed to perform tasks with the least human intervention, operator involvement is mandatory to ensure fault-free operations. Numerous technological solutions ...
详细信息
ISBN:
(纸本)9780791887240
Although Computer Numerical Control (CNC) machines were designed to perform tasks with the least human intervention, operator involvement is mandatory to ensure fault-free operations. Numerous technological solutions utilizing Artificial Intelligence, sensor fusion, Internet of Things (IoT), machine vision, etc., have been developed for process, component, and machine monitoring to impart smartness and autonomous operating abilities. The primary focus of these solutions is to monitor process faults such as tool wear, chatter, static deflections, and cutting forces to assist the operator in minimizing the consequences. The present work develops a vision-based solution for identifying uncommon process abnormalities like improper coolant flow, chip clogging, and tool breakage during CNC milling. The proposed solution replicates the task of a machine operator in identifying these faults and assists in fault-free operations. The study explores the feasibility of utilizing classical and deep learning-based object detection algorithms while developing these solutions. The classical image processing algorithm is ineffective during dynamic process conditions. The deep learning-based algorithm, with an average precision of about 0.75, showed proficiency in abnormalities detection. A Graphical User Interface (GUI) has been developed and integrated with the CNC milling machine to provide an interactive in-process monitoring tool. It is demonstrated that the proposed solution can reduce dependence on a machine operator while monitoring these faults.
暂无评论