Edge Computing provides mobile and Internet-of-Things (IoT) app vendors with a new distributed computing paradigm which allows an vendor to deploy its at hired edge servers near users the of cloud. This way, can be allocated nearby minimize network latency energy consumption. A cost-effective user allocation (EUA) requires maximum served minimum overall system cost. Finding centralized optimal solution this EUA problem is NP-hard. Thus, we propose EUAGame, game-theoretic approach that...

Background. Recent years have seen an increasing interest in cross-project defect prediction (CPDP), which aims to apply models built on source projects a target project. Currently, variety of (complex) CPDP been proposed with promising performance. Problem. Most, if not all, the existing are compared against those simple module size that easy implement and shown good performance literature. Objective. We aim investigate how far we really progressed journey by comparing between models....

With the widespread application of machine learning (ML) software, especially in high-risk tasks, concern about their unfairness has been raised towards both developers and users ML software. The software indicates behavior affected by sensitive features (e.g., sex), which leads to biased illegal decisions become a worthy problem for whole engineering community.

With the increasing application of deep learning (DL) models in many safety-critical scenarios, effective and efficient DL testing techniques are much demand to improve quality models. One major challenges is data gap between training construct evaluate them. To bridge gap, testers aim collect an subset inputs from contexts, with limited labeling effort, for retraining

Background. Self-admitted technical debt (SATD) is a special kind of that intentionally introduced and remarked by code comments. Those debts reduce the quality software increase cost subsequent maintenance. Therefore, it necessary to find out resolve these in time. Recently, many automatic approaches have been proposed identify SATD. Problem. Popular IDEs support number predefined task annotation tags for indicating SATD comments, which used projects. However, such clear prior knowledge...

Managing cross-project dependencies is tricky in modern software development. A primary way to manage using dependency configuration files, which brings convenience the entire ecosystem, including developers, maintainers, and users. However, developers may introduce smells if files are not well written maintained. Dependency recurring violations of management can potentially lead severe consequences. This paper provides an in-depth look at three smells, namely, <italic...

Deep learning (DL) models have proven to be highly successful and are now essential our everyday routines. However, DL models, like traditional software, inevitably contain bugs that affect their performance in real-world scenarios. Effective software engineering techniques necessary ensure dependability. In recent years, fault localization methods for gained significant attention as a valuable tool improving the reliability of models. Owing data-driven programming paradigm, challenging...

Life and death are inevitable processes in life, religions have unfolded rich imaginations discussions around this topic. Taoism ancient Greek religion, as of the East West respectively, formed their own unique views on life death. This paper, through literature analysis comparative analysis, studies reflected deities that govern compares similarities differences death, explores reasons behind them. The study finds believes human originates from Dao, while religion holds humans were created...

ABSTRACT The evolution (e.g., development and maintenance) of deep learning (DL) models has attracted much attention. One the main challenges during maintenance DL is model training, which often requires a lot human resources computing power (such as labeling costs parameter training). In recent years, to alleviate this problem, researchers have introduced idea software engineering (SE) into DL. Researchers consider new type software, borrowing practice traditional reuse, that is, focusing...

Endometriosis is a common chronic neuroinflammatory disease with poorly understood pathogenesis. Molecular changes and specific immune cell infiltration in the eutopic endometrium are critical to progression. This study aims explore mechanisms molecular differences proliferative of endometriosis by integrating bulk RNA-seq single-cell RNA sequencing (scRNA-seq) data, develop diagnostic predictive models for disease. Gene expression profiles from patients healthy controls were obtained...

ABSTRACT The relationship between test code and production code, that is, test‐to‐code traceability, plays an essential role in the verification, reliability, certification of software systems. Prior work on traceability focuses mainly Java. However, as Python allows more flexible testing styles, it is still unknown whether existing approaches well projects. In order to address this gap knowledge, paper evaluates can accurately identify links We collected seven popular projects carried out...

Machine learning (ML) software employs statistical algorithms to perform high-stake tasks in our daily lives, whose results are usually discriminatory due protected features (e.g., gender), i.e., one part (called privileged, e.g., male) may be more likely obtain beneficial decisions than the other unprivileged, female). In alleviating unfairness, developers have obtained widely-held beliefs about trade-off between performance and fairness for ML software. Surprisingly, recent research on...

Accurate software defect prediction could help practitioners allocate test resources to defect-prone modules effectively and efficiently. In the last decades, much effort has been devoted build accurate models, including developing quality predictors modeling techniques. However, current widely used such as code metrics process not well describe how change over project evolution, which we believe is important for prediction. order deal with this problem, in paper, propose use Historical...

In recent years, dynamic languages such as Python have become popular due to their flexibility and productivity. The lack of static typing makes programs face the challenges fixing type errors, early bug detection, code understanding. To alleviate these issues, PEP 484 introduced optional annotations for in 2014, but unfortunately, a large number are still not annotated by developers. Annotation generation tools can utilize inference techniques. However, several important aspects annotation...

Software build integrates modules developed and maintained by different developers in parallel, tests the result of integration, serves as a crucial step cooperatiive software development. Predicting has drawn interest academia industry. In spite many previous researches, generalizability failure prediction over wide range open-source projects remains unclear.In this paper, we used 9 classifiers to construct models investigated performance both cross-validation on-line predictions on 126...

The dynamic typing discipline of Python allows developers to program at a high level abstraction. However, type related bugs are commonly encountered in systems due the lack declaration and static checking. Especially, misuse produces underlying increases maintenance efforts. In this paper, we introduce six types practices programs, which common but potentially risky usage by developers. We also implement tool named PYDYPE detect them. Based on tool, conduct an empirical study nine...

Background. Code-line-level bugginess identification (CLBI) is a vital technique that can facilitate developers to identify buggy lines without expending large amount of human effort. Most the existing studies tried mine characteristics source codes train supervised prediction models, which have been reported be able discriminate code amongst others in target program. Problem. However, several simple and clear characteristics, such as complexity lines, disregarded current literature. Such...

The boom of DL technology leads to massive models built and shared, which facilitates the acquisition reuse models. For a given task, we encounter multiple available with same functionality, are considered as candidates achieve this task. Testers expected compare select more suitable ones w.r.t. whole testing context. Due limitation labeling effort, testers aim an efficient subset samples make precise rank estimation possible for these To tackle problem, propose Sample Discrimination based...