What are developers talking about information security? A large-scale study using semantic analysis of Q&A posts.

Background: Digitalization and rapid technological improvement in the present day bring numerous benefits, but they also raise the complexity and diversity of cyber security risks, putting critical information security issues on the agenda. Growing issues and worries about information security endanger not only the security of individuals and organizations but also global social and economic stability. Methods: This study investigates the issues and challenges regarding information security by analyzing all the postings on ISSE (Information Security Stack Exchange), a Q&A website focused on information security. In order to identify the primary topics addressed in postings shared on the ISSE platform, we employed a probabilistic topic modeling method called latent Dirichlet allocation (LDA), which is generative in nature and relies on unsupervised machine learning processes. Results: Through this investigation, a total of 38 topics were identified, demonstrating the present state of information security issues and challenges. Considering these topics, a comprehensive taxonomy of seven categories was devised to address information security issues, taking into account their backgrounds and perspectives. Subsequently, we conducted an examination of the prevalence and complexity of the matters at hand. In addition, we have defined the prevailing technologies utilized in the realm of information security, including tasks, certifications, standards, methods, tools, threats, and defenses. We have provided a number of implications for different stakeholders, including academics, developers, educators, and practitioners, who are working towards advancing the field of information security.

Identification of mobile development issues using semantic topic modeling of Stack Overflow posts.

Background: Increasing demands for mobile apps and services have recently led to an intensification of mobile development activities. With the proliferation of mobile development, there has been a major transformation in the architectures, paradigms, knowledge domains and skills of traditional software systems towards mobile development. Therefore, mobile developers experience a wide spectrum of issues specific to development processes of mobile apps and services. Methods: In this article, we conducted a semantic content analysis based on topic modeling using mobile-related questions on Stack Overflow, a popular Q&A site for developers. With the aim of providing an understanding of the issues and challenges faced by mobile developers, we used a semi-automated methodology based on latent Dirichlet allocation (LDA), a probabilistic and generative approach for topic modeling. Results: Our findings revealed that mobile developers' questions focused on 36 topics in six main categories, including "Development", "UI settings", "Tools", "Data Management", "Multimedia", and "Mobile APIs". Besides, we investigated the temporal trends of the discovered issues and their relationships with mobile technologies. Our findings also revealed which issues are the most popular and which issues are the most difficult for mobile development. The methodology and findings of this study have valuable implications for mobile development stakeholders including tool builders, developers, researchers, and educators.

What issues are data scientists talking about? Identification of current data science issues using semantic content analysis of Q&A communities.

Background: Because of the growing involvement of communities from various disciplines, data science is constantly evolving and gaining popularity. The growing interest in data science-based services and applications presents numerous challenges for their development. Therefore, data scientists frequently turn to various forums, particularly domain-specific Q&A websites, to solve difficulties. These websites evolve into data science knowledge repositories over time. Analysis of such repositories can provide valuable insights into the applications, topics, trends, and challenges of data science. Methods: In this article, we investigated what data scientists are asking by analyzing all posts to date on DSSE, a data science-focused Q&A website. To discover main topics embedded in data science discussions, we used latent Dirichlet allocation (LDA), a probabilistic approach for topic modeling. Results: As a result of this analysis, 18 main topics were identified that demonstrate the current interests and issues in data science. We then examined the topics' popularity and difficulty. In addition, we identified the most commonly used tasks, techniques, and tools in data science. As a result, "Model Training", "Machine Learning", and "Neural Networks" emerged as the most prominent topics. Also, "Data Manipulation", "Coding Errors", and "Tools" were identified as the most viewed (most popular) topics. On the other hand, the most difficult topics were identified as "Time Series", "Computer Vision", and "Recommendation Systems". Our findings have significant implications for many data science stakeholders who are striving to advance data-driven architectures, concepts, tools, and techniques.

Comparison of Radiation-Induced Secondary Malignancy Risk Between Sequential and Simultaneous Integrated Boost for the Treatment of Nasopharyngeal Carcinoma: Intensity-Modulated Radiotherapy versus Volumetric-Modulated Arc Therapy.

PURPOSE: This study aimed to compare the secondary cancer risk (SCR) between the sequential boost (SEQ) technique and simultaneous integrated boost (SIB) technique in intensity-modulated radiotherapy (IMRT) and volumetric-modulated arc therapy (VMAT) in patients with nasopharyngeal carcinoma (NPC) using the concepts of organ equivalent dose (OED) and excess absolute risk (EAR). PATIENTS AND METHODS: IMRT-SEQ, VMAT-SEQ, IMRT-SIB, and VMAT-SIB plans were created with identical objective functions for five patients with early-stage NPC. Three different planning tumor volumes (PTVs; PTV1, PTV2, and PTV3) were delineated for each patient, and the prescribed doses were 50 Gy, 60 Gy, and 70 Gy (2 Gy/fraction), respectively, for the SEQ technique and 52.8 Gy, 59.4 Gy, and 69.3 Gy (33 fractions), respectively, for the SIB technique. RESULTS: All plans were clinically acceptable. There was no difference in most OED-based SCRs between IMRT and VMAT when the same fractionation scheme was used. Compared with the SEQ technique, the SIB technique in IMRT and VMAT was associated with the lowest OEDs for the oral cavity, pharynx, parotids, and submandibular glands, resulting in SCR reduction. SCR for the parotids was much lower than that for the other assessed organs when the SIB technique was used. CONCLUSION: Our findings suggest that OED-based SCRs are lower with the SIB technique than with the SEQ technique in IMRT and VMAT in most organs for which SCR was calculated; furthermore, SCR for the parotids is much lower than that for other organs when the SIB technique is used in patients with NPC.

Secondary cancer risk after whole-breast radiation therapy: field-in-field versus intensity modulated radiation therapy versus volumetric modulated arc therapy.

OBJECTIVE: In this study, we used the concept of organ-equivalent dose (OED) to evaluate the excess absolute risk (EAR) for secondary cancer in various organs after radiation treatment for breast cancer. METHODS: Using CT data set of 12 patients, we generated three different whole-breast radiation treatment plans using 50 Gy in 2 Gy fractions: three-dimensional conformal radiotherapy with a field-in-field (FinF) technique, intensity modulated radiation therapy (IMRT), and volumetric modulated arc therapy (VMAT). The OEDs were calculated from differential dose-volume histograms on the basis of the "linear-exponential," "plateau," and "full mechanistic" dose-response models. Secondary cancer risks of the contralateral breast (CB), contralateral lung (CL), and ipsilateral lung (IL) were estimated and compared. RESULTS: The lowest EARs for the CB, CL, and IL were achieved with FinF, which reduced the EARs by 77%, 88%, and 56% relative to those with IMRT, and by 77%, 84%, and 58% relative to those with VMAT, respectively. The secondary cancer risk for FinF was significantly lower than those of IMRT and VMAT. OED-based secondary cancer risks for CB and IL were similar when IMRT and VMAT were used, but the risk for CL was statistically lower when VMAT was used. CONCLUSION: The overall estimation of EAR indicated that the radiation-induced cancer risk of breast radiation therapy was lower with FinF than with IMRT and VMAT. Therefore, when secondary cancer risk is a major concern, FinF is considered to be the preferred treatment option in irradiation of whole-breast. ADVANCES IN KNOWLEDGE: Secondary malignancy estimation after breast radiotherapy is becoming an important subject for comparative treatment planning.When secondary cancer risk a major concern, FinF technique is considered the preferred treatment option in whole breast patients.