Please read below student posts reply each in 125 words.
Ramesh – Decision tree
A tree has numerous real-life analogies and has influenced a broad range of machine learning that covers regression and classification. A decision tree can be used to visualize decisions and decision making explicitly and clearly. It uses a tree-like decision-making model, as the name goes. Although it is a common tool to develop a strategy for achieving a specific goal in data mining, this is also common in machine learning world (Yu, Haghighat, Fung, & Yoshino, 2010). Some of the very basic example is using it to predict whether or not a passenger survives, using titanic data.
A decision tree with its base on the top is drawn upside down. The end of the not-divided section is the final decision and based on the titanic example, is whether the passenger died or survived. While there is more functionality for an individual dataset and this is only a branch in a bigger tree, the simplicity of the algorithm cannot be overlooked (Yu, Haghighat, Fung, & Yoshino, 2010). The value of the feature is apparent and relationships can be easily seen. This method is commonly known as the learning decision tree, and above the tree. From the example, the tree is called classification tree because it is intended to determine the passenger as being survived or dead.
The only difference between Regression trees and decision tree is that they only predict constant values like the house price. Generally, CART algorithms or classification and regression trees are referred to by Decision Tree algorithms (Yu, Haghighat, Fung, & Yoshino, 2010). One importance of decision tree is that it can help a manager or financial officer make decisions faster since the summary of the information he or she needs to make decision is visually represented. The disadvantage is that the tree may contain numerous branches for a large dataset with numerous decisions.
Pranay – Random forests:
Random forest adapts multiple classification trees to a set of data and then combines predictions for all trees from different versions. Instead of attempting to predict the entire dataset, and then combine it all into one classification, the prediction is based only on the main features of the dataset. By taking multiple small subsets of the data (e.g. independent variables from multiple datasets or data-like/data-like overlapping items from various data sources) and making a prediction from them, there is no need for gradient descent, edge-based methods, tree-based methods or other expensive tasks. (Reddy, A. M. (2020))
The value of model comparison is not to create a trained model from scratch, but to predict the prediction from another model (using a comparison method). By a problem of L1 loss, ROC curves, validation statistics and confidence intervals, he expects a reasonable fit of the data in his data based setting with some margin for errors. The objective of this approach is to identify best fit models that preserve or cause the best predictions in specific models. My approach is somewhat different than that of Schimpf. They use an objective validation in a development environment, where their options are restricted to a model that takes over from any other model without additional tweaking. In my case, a comparison method with inference constraints requires me to explicitly check for internal consistency, by proposing different sources of changes within the data. We can use multiple hypotheses for models 21. They can find a useful model for their data in other words, their models need to make sense in the data in relatively few instances if all components of the model are selected that match the data as closely as possible. Such models can be a useful starting point for customizations such as extra parameters, connections, layers. (Nettleton, D. (2020))
saigutta – Cyber-Attack
The advancement of cyber-attacks has raised different contentions. Currently, the pace of cyber-attacks is more than any other type of crime. Cyber-security is a worldwide issue influencing numerous enterprises. Although various legislatures have set up a few countermeasures, cyber-attacks appear to be expanding every year. A cyber-attack on pipeline industries in the United States showcase excellent proof of a security concern that may influence the economy of the nation.
Hackers though their various sophisticated strategies, concentrate on more than one explicit organization. All organizations are helpless against digital attacks. Pipeline companies in the US fell in a comparative snare in 2002. The cyberattack happened after UglyGorilla took private data from pipeline enterprises (Northey, Sobczak, and Behr, 2017). Such attacks accompany a remarkable effect on the national economy. Numerous businesses both private and public, rely upon gases for different operations. A disruption in pipeline ventures in any event, for a brief timeframe, implies a vast decrease in electric force that could bring about impressive misfortunes in the economy. Notably, Pipeline industries have huge shares in foreign exchange. Thus, a slight interruption can cause the economy of the United States to decline significantly.
Many organizations, besides pipeline sectors, are vulnerable to cyber-attacks because they deploy weak security measures. It is fundamental for pipeline industries to identify various threats and be aware of multiple mechanisms that hackers use to conduct cyber â€“attacks (Bhardwaj & Goundar, 2018). However, identifying and acknowledging cybercrimes can only succeed by using a diversity strategy. Typically, diversity entails incorporating all knowledge of cyber-attacks from all people and nations. Several people have crucial information that can help organizations eliminate and mitigate cyber securities. However, some companies many never consider their views due to differences in age, race, ethnicity among others. Utilizing diversity strategy can yield significant impacts in the cyber-attacks world.
Nagireddy – Cyber-Attack in Pipeline Industry
Natural gas and crude oil industries have become more vulnerable to cyber-attacks. The risks of cyber-attacks have increased because many pipeline industries use old operation technology as well as insecure industrial control systems hence making them lack essential protection strategies such as network segmentation and passwords. Moreover, hackers have advanced their cyber-attacks skills. They deploy more sophisticated techniques that make all industries vulnerable to cybersecurity threats. When a cyber-attack hit one of the largest pipeline industries five years ago, it demonstrated the potential risks that could occur in future.
The cyber-attack pipeline industry remains one of the unforgettable incidences in the United States. The attack highlights the risk that all industries suffer regardless of the many countermeasures implemented. A hacker by the name UglyGorilla access and trove confidential data of pipeline industries, breaching approximately 300,000 steel web that was fundamental in the US economy (Northey, Sobczak & Behr, 2017). since the attack, increased dependence of natural gas to generate power has made the oil transmission system to become consequential targets for hacking in the nation. Notably, UglyGorilla still holds some confidential information that can launch another cyber-attack in future.
Besides another cyber-attack, the number of advanced threats has increased after UglyGorilla alleged the intrusion. The increasing level of the attack appears to disrupt the economy of the US by affecting gas transmission and electric power generation. Affecting the pipeline industry means influencing security and exchange that lower the economy (Northey, Sobczak & Behr, 2017). Moreover, transportation and manufacturing industries may stop their operations that in turn, affect the economy at higher margins.
Although governments and pipeline industries have implemented several countermeasures to reduce cyber-attack, they should utilize diversity strategy as well to reduce the security threat. Typically, diversity means using ideas from all stakeholders regardless of ethnicity, race, religion, gender, and sexual orientation (Okhravi & Shrobe, 2017). Moreover, diversity entails obtaining help from other nation that appears to be successfully mitigating the cyber-attack threats.
satya – A cyber threat that has the biggest impact on the economy uses the most sophisticated or criminal techniques. In recent years, we have seen a rapid increase in the use of targeted cyberattacks by nation-states to cripple critical industries and electrical grids, send spoofed emails, and otherwise manipulate markets and peoples’ behaviors. In order to fully understand the global implications of targeted attacks, the data necessary for forensics and risk analysis must be cross-linked and analyzed, in particular, when investigating commonality (Darden, 2016). While this would seem a task for our best and brightest, this reality is a sharp reminder of our society’s limitations. A cyber threat that has the biggest impact on the economy is likely to have the most complex computer network and, to the extent that the individuals involved and the companies they work for have differing levels of digital protection, differences in information processing and security, the effectiveness of their security measures, etc., these differences, too, will ultimately add up to a substantial increase in economic costs. Where the precise amount of cost is hard to ascertain, however, we believe it is important to consider the impact on the global economy in deciding how best to mitigate those costs, if we are to ensure that our efforts do not harm our nation’s economic well-being and social cohesion. In part, this will depend on the appropriate methodology for defining an economic cost. The problems go beyond how vulnerable the Internet is to corporate hacking. They also go beyond the implications of a network’s fragility and the effectiveness of the sharing of information required to build common defenses. First, network architecture and protocols may be inadequate for attaining the level of interoperability required to confront high-impact cyber threats (Mylonas, 2016).
A unique attack vector that will surprise and overload both the highly automated security tools used by the intelligence community and the banks and financial institutions using the tools that exist in the world. Security through obscurity is no longer an option for defenders. By implementing new, more resilient, cyber defenses, Wall Street institutions will start to perceive themselves as having new levels of vulnerability that they can better protect (Ahn, 2016). The perception of an institution’s risk-reward ratio shifts from growing protection to becoming more exposed. Given that most cyber threats to the financial sector are undetectable or hard to reverse engineer. Economy and society can be categorized according to the type of damage it causes. Each type of damage, however, has a unique form of impact. When targeting an individual, cyber intruders might attempt to gain access to personal information, take down personal financial records, or put out malicious advertising. Once the intents are complete, a cyber-attack could cause huge losses to businesses