SHAN_KAR_PUTH_NAG_REPLIES
Name:
Date:
Institution:
Reply to Hirihara
You make an interesting point, first, in regard to office design as impacted by growth in telecommunication, for instance, the decrease in designated desks, smaller office spaces, and smaller conference rooms. However, I do not see the need for a lounge space within an office; I would argue that such a facility would encourage employees to be present at work as a result of comfort provided therefore working against the carbon footprint reduction (Gap, 2003). Employees would be required to work from home. Second, I totally concur with your assertions regarding the reduced use of office phones as a result of the proliferation of telecommunication software as well as the corresponding supportive hardware (Pier et al., 2017)
Reply to Santosh
I am amazed by these assertions. The reasoning that software implementation costs will influence the approach used for the process. The argument that the globe will experience significant growth in mobile and telecommunication. The pose that the agent-based programs will be the way to go in the next decades in order to reduce overheads. And, the deployment of the active modeling approach for managing functional complexities reveals coherence of thought. I also concur that care should be taken to manage the demerits that would arise as a result of the use of the aforementioned technologies.
Reply to Divya
Data mining is a complex process, and I appreciate the techniques you outlined for this process. It is important to point out that while you describe the rule-based method, nearest neighbor approach, Naïve Bayes and Bayesian Belief Networks, Memory Based Reasoning, and support vector machine, it is important to point out the relationship between these processes. One would ask, where does this description meet the decision trees, neural networks as well as Linear and Logistic regression? It would be difficult to tell, as these are also considered classification models.
Reply to Hersha
I would agree to a larger extent with the descriptions you have nade regarding made regarding the elective strategies, including rule-based, probabilistic, deterministic, direct, worldwide, and generative approaches of classification. I am particularly impressed with the arguments regarding the impacts of non-linear models on the worldwide classifier. You argue that “this one-size-fits-all system may not be compelling when the connection between the traits and the class marks shifts over the input space (Almarabeh, 2017),” This assertion is very significant.
Reply to Neha
The arguments fronted in his post are interesting. First, Neha points out the criticality of information to the modern-day organization. Further, it is evident in the post that in order to gain useful benefits from information, there is a need for proper information gathering and grouping. Second, I agree with the manner in which the arguments are made, the author compares the classifiers, from these arguments, it is easy to point out the merits of support vector machine (SVM) approaches over the decision trees, rule-based as well as the standard-based models. Further, I concur with the assertion that neural networks are valuable for administered learning and non-regulated learning while the Bayesian models of classification depend on a Bayesian network that addresses joint probability throughout a ton of straight out characteristics (Wang et al., 2018). I also point out that it is important to describe the challenges of such approaches, for instance, the impact of bad input selection, noise data, insufficient stop condition among other challenges associated with neural networks and the challenge of creating node probability tables (NPT) in Bayes Network (Almarabeh, 2017).
Reply to Prathiba
Prathiba makes interesting points while comparing the data mining classifiers. The author shows the similarities between rule-based classifieds, the standard-based classifiers, and decision trees. Further, Prathipa asserts that these classifiers make rectilinear space apportioning and allot a class to each parcel. However, the standard-based approach has an advantage over the other aforementioned classifiers because it can permit various principles to be actuated for a given occasion, permitting more perplexing models to be scholarly. The advanced arguments concerning rule-based classifiers, RVM, and Bayesian are true. For instance, I would agree with the point that Rule-based classifiers are not appropriate for dealing with missing qualities in the test set (Wang et al., 2018).
Reply to Pravin
The post describes the cloud computing infrastructure. From the post, it is clear that the cloud infrastructure is beneficial to the modern-day persons as well as organizations. It reduces overheads as a result of fewer purchases of resources in terms of hardware or hardware as these are taken care of by a cloud service provider. I am particularly interested in the description of the cloud infrastructure that constitutes a logical network perimeter, a cloud usage monitor, a resource replication system, a virtual server, a cloud storage device, and a ready-made environment. In my opinion, I would have used terms like management software, deployment software, hypervisor, a network, and a server (Raj et al., 2009) because of the simplicity and universality of those terms. Further, Parvin argues that the virtual network parameter is useful for establishing a boundary within a network that isolates specific IT resources for physical distribution (Zhang et al. 2012), this is a significant concept in cloud computing.
Reply to Rahul
Rahul offers an appropriate definition to cloud computing, outlining the fact that the infrastructure employs remote services that are internet hosted to store, process, manage, and distribute data that support critical activities. To a large extent, I agree with this point; however, concerning the aspects of cloud computing, the author fails to mention these dimensions. These aspects could include broad network access, measured service, on-demand service, multi-tenancy and resource pooling, and rapid elasticity and scalability (Raj et al., 2009). In regard to cloud characteristics and mechanisms. I largely agree with Rahul’s arguments. While in mechanism and characteristics, the cloud infrastructure meets the five aspects (Zhang et al. 2012).
References
Almarabeh, H. (2017). Analysis of students’ performance by using different data mining classifiers. International Journal of Modern Education and Computer Science, 9(8), 9.
Gap, W. (2003). Flexible work arrangements.
Raj, H., Nathuji, R., Singh, A., & England, P. (2009, November). Resource management for isolation enhanced cloud services in Proceedings of the 2009 ACM workshop on Cloud computing security (pp. 77-84).
Pier, E. L., Raclaw, J., Ford, C. E., Kaatz, A., Carnes, M., & Nathan, M. J. (2017). Videoconferencing in Peer Review: Exploring Differences in Efficiency and Outcomes. Philadelphia, PA: International Society of the Learning Sciences.
Wang, F., Wang, Q., Nie, F., Yu, W., & Wang, R. (2018). Efficient tree classifiers for large scale datasets. Neurocomputing, 284, 70-79.
Zhang, M., Ranjan, R., Haller, A., Georgakopoulos, D., Menzel, M., & Nepal, S. (2012, October). An ontology-based system for Cloud infrastructure services’ discovery. In 8th international conference on collaborative computing: networking, applications, and work sharing (CollaborateCom) (pp. 524-530). IEEE.