Xai4fl

Widyasari, Ratnadira; Prana, Gede Artha Azriadi; Haryono, Stefanus Agus; Tian, Yuan; Zachiary, Hafil Noer; Lo, David

doi:10.1145/3524610.3527902

Cited by 9 publications

(4 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the future, we implement additional methods to further improve the effectiveness of fault localization in student programs. In future work, we will do more research further improve the fault localization accuracy of student programs, which includes but is not limited to: (1) using interpretable machine learning methods for fault localization [29]. (2) using causal inference to improve fault localization, and (3) using students' historical version program information for fault localization.…”

Section: Discussionmentioning

confidence: 99%

Boosting Spectrum-Based Fault Localization via Multi-Correct Programs in Online Programming

ZHENG,

HU,

CHEN

et al. 2024

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

Providing students with useful feedback on faulty programs can effectively help students fix programs. Spectrum-Based Fault Location (SBFL), which is a widely studied and lightweight technique, can automatically generate a suspicious value of statement ranking to help users find potential faults in a program. However, the performance of SBFL on student programs is not satisfactory, to improve the accuracy of SBFL in student programs, we propose a novel Multi-Correct Programs based Fault Localization (MCPFL) approach. Specifically, We first collected the correct programs submitted by students on the OJ system according to the programming problem numbers and removed the highly similar correct programs based on code similarity, and then stored them together with the faulty program to be located to construct a set of programs. Afterward, we analyzed the suspiciousness of the term in the faulty program through the Term Frequency-Inverse Document Frequency (TF-IDF). Finally, we designed a formula to calculate the weight of suspiciousness for program statements based on the number of input variables in the statement and weighted it to the spectrum-based fault localization formula. To evaluate the effectiveness of MCPFL, we conducted empirical studies on six student program datasets collected in our OJ system, and the results showed that MCPFL can effectively improve the traditional SBFL methods. In particular, on the EXAM metric, our approach improves by an average of 27.51% on the Dstar formula.

show abstract

Section: Discussionmentioning

confidence: 99%

Boosting Spectrum-Based Fault Localization via Multi-Correct Programs in Online Programming

ZHENG,

HU,

CHEN

et al. 2024

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

show abstract

“…The limitations of identifying tracking code share similarities with prior research on fault localization. For example, spectra-based fault localization (SBFL) [43,45,65,67,72,77] leverages the statement coverage using the set of passing and failing test cases to localize the statement that is most likely to induce a test failure. Similarly, Bela et al [75] and Laghari et al [52] present an approach that uses the frequency of method occurrence in the call stack of failing test cases for localizing the faulty methods.…”

Section: Related Workmentioning

confidence: 99%

Blocking JavaScript without Breaking the Web: An Empirical Investigation

Amjad¹,

Shafiq²,

Gulzar³

2023

Preprint

View full text Add to dashboard Cite

Modern websites heavily rely on JavaScript (JS) to implement legitimate functionality as well as privacy-invasive advertising and tracking. Browser extensions such as NoScript block any script not loaded by a trusted list of endpoints, thus hoping to block privacy-invasive scripts while avoiding breaking legitimate website functionality. In this paper, we investigate whether blocking JS on the web is feasible without breaking legitimate functionality. To this end, we conduct a large-scale measurement study of JS blocking on 100K websites. We evaluate the effectiveness of different JS blocking strategies in tracking prevention and functionality breakage. Our evaluation relies on quantitative analysis of network requests, and resource loads as well as manual qualitative analysis of visual breakage. First, we show that while blocking all scripts is quite effective at reducing tracking, it significantly degrades functionality on approximately two-thirds of the tested websites. Second, we show that selective blocking of a subset of scripts based on a curated list achieves a better tradeoff. However, there remain approximately 15% "mixed" scripts, which essentially merge tracking and legitimate functionality and thus cannot be blocked without causing website breakage. Finally, we show that fine-grained blocking of a subset of JS methods, instead of scripts, reduces major breakage by 3.7× while providing the same level of tracking prevention. Our work highlights the promise and open challenges in fine-grained JS blocking for tracking prevention without breaking the web.

show abstract

“…The limitations of identifying tracking code share similarities with prior research on fault localization. For example, spectra-based fault localization (SBFL) [33,42,62,64,74] leverages the statement coverage using the set of passing and failing test cases to localize the statement that is most likely to induce a test failure. Similarly, Bela et al [71] and Laghari et al [48] present an approach that uses the frequency of method occurrence in the call stack of failing test cases for localizing the faulty methods.…”

Section: Related Workmentioning

confidence: 99%

Blocking JavaScript Without Breaking the Web: An Empirical Investigation

Amjad

Shafiq

Gulzar

2023

PoPETs

View full text Add to dashboard Cite

Modern websites heavily rely on JavaScript (JS) to implement legitimate functionality as well as privacy-invasive advertising and tracking. Browser extensions such as NoScript block any script not loaded by a trusted list of endpoints, thus hoping to block privacy-invasive scripts while avoiding breaking legitimate website functionality. In this paper, we investigate whether blocking JS on the web is feasible without breaking legitimate functionality. To this end, we conduct a large-scale measurement study of JS blocking on 100K websites. We evaluate the effectiveness of different JS blocking strategies in tracking prevention and functionality breakage. Our evaluation relies on quantitative analysis of network requests and resource loads as well as manual qualitative analysis of visual breakage. First, we show that while blocking all scripts is quite effective at reducing tracking, it significantly degrades functionality on approximately two-thirds of the tested websites. Second, we show that selective blocking of a subset of scripts based on a curated list achieves a better trade-off. However, there remain approximately 15% “mixed” scripts, which essentially merge tracking and legitimate functionality and thus cannot be blocked without causing website breakage. Finally, we show that fine-grained blocking of a subset of JS methods, instead of scripts, reduces major breakage by 3.8× while providing the same level of tracking prevention. Our work highlights the promise and open challenges in fine-grained JS blocking for tracking prevention without breaking the web.

show abstract

Xai4fl

Cited by 9 publications

References 31 publications

Boosting Spectrum-Based Fault Localization via Multi-Correct Programs in Online Programming

Boosting Spectrum-Based Fault Localization via Multi-Correct Programs in Online Programming

Blocking JavaScript without Breaking the Web: An Empirical Investigation

Blocking JavaScript Without Breaking the Web: An Empirical Investigation

Contact Info

Product

Resources

About