Artificial Intelligence and Building Information Modeling Software Tools Research Paper

My discipline is Networking and I have chosen two articles related to the modern technology. The first article is written by Benzaid and Taleb (2020) published in IEEE Network, and it examines how artificial intelligence influence 5G network. The second article is by Zhang et al. (2021) and it analyzes the interoperability of BIM software tools and addresses the problems in the process of data exchange. It was published in Computer Applications in Engineering Education journal. In addition to researchers, the audience of the first article is data security agents as it provides some insights on how the use of artificial intelligence to 5G can be venerable to external attacks. The second article’s audience is educators in engineering education as it discusses about the method of teaching engineering. This paper examines articles of Benzaid and Taleb (2020) and Zhang et al. (2021).

To begin with, there is a need to understand the content of each article. Both articles are scientific and made by the use of quantitative and qualitative research methods. Benzaid and Taleb (2020) warn the audience that to fully reap the benefits of 5G, it is critical to develop robust and long-lasting security mechanisms that can deal with the ever-changing threat landscape. They suggest that traditional security measures are not enough. This is because of given the growing number of vulnerabilities, the sophistication of cyber threats, the high volume of traffic, and the diverse technologies (e.g., SDN, NFV) and services that will shape next-generation wireless networks. According to Benzaid and Taleb (2020), a new security measure that needs to be considers is the adoption of Artificial Intelligence (AI). They describe AI as a method that would enable intelligent, adaptive, and autonomous security management, allowing for prompt and cost-effective detection and mitigation of security threats. Their description provides an impression that AI is a promising direction. Moreover, Benzaid and Taleb (2020) illustrate AI’s positive aides, such as its ability to identify hidden patterns in a vast set of time-varying multi-dimensional data that allow faster and more accurate decision-making.

With regards to the second article of Zhang et al. (2021), it reports a case study of the development of the new capstone project for engineering major students. It is based on team-based learning (TBL) combined with the 360-degree evaluation feedback method to increase students’ BIM competency. Data is collected and analyzed using a mix of qualitative and quantitative methodologies in order to assess students’ learning outcomes and BIM competency. The findings show that TBL, when combined with 360-degree feedback in the capstone project, can significantly improve graduates’ BIM expertise. This research examines the interoperability of BIM software platforms, data sharing issues, and recommendations for improving the course and BIM team collaboration. Compared to the first article, the second one is based on a study and has significant evidence to support their claim. The article of Benzaid and Taleb (2020) appeals to the audience by considering all potential security scenarios and analyzing specific aspects of AI and 5G network.

The study of Zhang et al. (2021) discovered how students’ professional capabilities may be increased through a capstone project, and educators can use the BIM course to build engineers that closely match industry needs. The paper makes a case for using the capstone project to help engineering students improve and cultivate their BIM proficiency in MEP systems. This research established a new paradigm for using TBL and 360-degree feedback in engineering education. In terms of the support for each article’s claims, Zhang et al. (2021) is more reliable as it has a specific case.

In terms of the format of both articles, they are both heavily theoretical, meaning that there are numerous citations and the use of previous literature. Benzaid and Taleb (2020) provide images of how AI and 5G work, interpreting their explanation. It was done to make easier understanding for the audience. For example, they illustrate mapping of the adversarial ML attacks to the ML5G high-level architecture. Meanwhile, Zhang et al. (2021) provide descriptive statistics and t test results as a table while making a comprehensive analysis of the data. As such, it can be assumed that both articles rely on logos rather than pathos as they are written scientifically, meaning that they are based on facts. In both articles, there was also a logical progression of ideas and claims that were supported by a great variety of numerical facts and evidence.

The importance of AI in encouraging security in 5G and beyond networks was highlighted in Benzaid and Taleb (2020) article. Meanwhile, it emphasized the security concerns that may accompany the anticipated AI benefits if unscrupulous actors take use of their potential or flaws. We advised many defense measures while advising on which components of the ML5G unified architecture they may be implemented in order to increase resilience to AI attacks. Despite their merits, each of the recommended countermeasures has its own set of limitations, and none of them can be considered an all-in-one solution for dealing with all AI dangers. As a result, one possible research avenue is to investigate how those countermeasures could be combined to satisfy both security and performance goals.

Zhang et al. (2021) are trying to convince their audience that applying TBL and 360-degree evaluation feedback is suitable for engineering education. If Industry 4.0 is to be effectively embraced across the construction industry, enhanced BIM competencies for the AECO sector are critical; engineering educational programs play a significant role in achieving this important goal. The unique combination of TBL and 360-degree evaluation feedback as part of the engineering capstone project plays a vital role in developing, evaluating, and acquiring educational capacity in undergraduate engineering majors, according to this empirical research study. The relationship between 360-degree assessment feedback and TBL integration to enable BIM competences in MEP systems for engineering students is the subject of this study, which is the first of its kind. The findings of the case study analysis show that BIM competencies are developed through all-around evaluation feedback and practice. The findings also show that collaborative ability has a significant impact, particularly in the cycle of evaluation, feedback, and improvement through reflection, which is critical for students’ BIM competency development.

The contents of both articles provide a comprehensive approach on their topics. The goal of articles is to inform the audience about their findings. The articles also want to convince the audience in importance of their work. The organization of articles is similar to each other, yet Benzaid and Taleb (2020) article has many sub-topics that help readers to understand the text properly. For example, there are various sb-topics related to 5G network management and risk assessment. The paragraphs of both articles are long and contain factual data along with their analysis. Each paper provides in depth literature review that states about the state of scholarship for the current time and some insights on aspects that were addressed in the literature. These previous scholarly works were cited by the use of Harvard citation style. From a reader’s perspective, the chosen citation style was suitable and convenient to check the sources.

The used methodology in both papers differs as they examine substantially different topics. Zhang et al. (2020) research study’s framework is separated into five sections. The study started with a look at BIM education and capstone projects in MEP systems, as well as TBL and 360-degree evaluation feedback in education. Second, in the BIM capstone, the TBL educational approach was devised, which was integrated with 360-degree evaluation feedback. This BIM capstone project approach (i.e., TBL paired with BIM 360-degree evaluation feedback) seeks to assist students and teachers through MEP systems teaching and practice, and its teaching content can be tailored to various engineering specialties’ needs. Third, TBL was paired with 360-degree evaluation feedback in the BIM capstone, followed by the use of 360-degree evaluation feedback to evaluate students’ learning outcomes and BIM competency, using a university in China as an example. Finally, the students’ teamwork, professional qualities, and BIM expertise were discussed, as well as the benefits of the BIM capstone project and suggestions for improvement.

The experimental research method was used in this work, and a case study method was used to describe the BIM capstone project in detail. The goal is to use the new capstone project in conjunction with TBL and 360-degree evaluation feedback as an intervention measure to see how it affects BIM capability. It was not possible to construct a control group due to the small number of students in this major, the same learning resources and environment for all participants, and the need to complete the requirements of the graduation audit and graduate easily. As a result, we used a series of experimental designs for comparative testing to determine the impact of the new capstone program’s implementation. The comparison of the experimental design’s learning effect aids in the exploration of the potential usefulness of new teaching techniques and curriculum innovation, which is appropriate for this study’s experimental setting. A semi-structured interview, evaluation criteria for capstone results, capstone scores and comparative test scores, and a review of the capstone process logbook are all part of the data gathering process for this study.

With regards to Benzaid and Taleb (2020) study, unlike Zhang et al. (2021), they provide potential actions of attackers to 5G network through the use of AI. Following that, they also demonstrate several defense mechanisms of how AI can be sued to prevent or even stop the attackers’ mechanisms. For example, defensive distillation is a training approach that employs the knowledge learned from a machine learning model to improve the model’s resilience to hostile cases. Both adversarial training and defensive distillation conduct implicit gradient masking, which consists of making the model’s gradient worthless by altering its direction or setting it to zero. Indeed, the lack of an actual gradient makes the development of adversarial cases more difficult, allowing the model to be more robust. This does not rule out the possibility that the model would be sensitive to adversarial samples created via transferability-based black-box attacks. Furthermore, the increased robustness provided by adversarial training and defensive distillation comes at the expense of reduced accuracy on clean data.

To create a robust model, ensemble methods mix multiple models. Ensemble approaches improve the model’s robustness while also raising its accuracy on clean samples. The advantage of ensemble approaches, however, comes at the cost of greater model complexity and processing cost. Defense Before feeding them into the ML model, GANs try to denoise input samples from adversarial perturbations by projecting them on to the range of the GAN’s generator. In other words, they want to select the sample that comes closest to the adversarial example that the GAN’s generator can generate and feed it into the ML model. Benzaid and Taleb (2020) also present some types of attack that can be mitigated by above-mentioned methods. This allows readers to fully understand the aspects of AI and 5G related problems. For example, the inference stage is the target of an evasive attack. These attacks, unlike poisoning attacks, have no effect on the training process. By adding minor perturbations to the input instances, the attacker attempts to escape the learnt model at test time. Adversarial examples are a type of perturbation.

To sum up, the papers of Benzaid and Taleb (2020) and Zhang et al. (2021) rely on logos when addressing their topics. The goal of each article was to inform and convince their audience regarding the importance of their findings. Both papers are heavily theoretical and follow a logical progression of ideas and claims that are supported by factual evidence and statistics. They also include potential outcomes and conditions that allow readers to fully engage with the content of articles.

References

Benzaïd, C., & Taleb, T. (2020). AI for beyond 5G networks: a cyber-security defense or offense enabler?. IEEE Network, 34(6), 140-147.

Zhang, J., Zhang, Z., Philbin, S. P., Huijser, H., Wang, Q., & Jin, R. (2021). Toward next‐generation engineering education: A case study of an engineering capstone project based on BIM technology in MEP systems. Computer Applications in Engineering Education.