Analytics final exam
Degree, closeness, and betweeness all fall into which one of the following categories? Select one: A. Centrality measures Degree, closeness, and betweeness all fall into the category of centrality measures. B. Matrices C. Sociograms D. Reputational risks
A
Computer output simulating human reasoning best describes which one of the following? Select one: A. Artificial intelligence B. Machine learning C. Deep learning D. Chatbots
A
A traditional workplace accident analysis technique that allows an organization to discern the cause of a harmful event and prevent such an event from recurring by identifying and mitigating the actions, inactions, conditions, or behaviors that caused the harmful event is Select one: A. Root cause analysis. B. System safety. C. Fault tree analysis. D. Failure mode and effects analysis.
A
After a predictive model is developed to evaluate estimates of ultimate losses Select one: A. The analysis should be repeated to determine accuracy. B. The model will remain valid for new claims. C. A different data analysis technique should test the model. D. There will be consistently accurate results in the future.
A
Annette is a claims adjuster with Midwestern Insurance (MI). MI recently invested in and established robotic process automation (RPA) in many of its routine repetitive tasks in its claims department. Annette is discovering that her role may be changing in which one of the following ways? Select one: A. Annette will assume responsibility for claims that call for empathy to meet the emotional needs of insureds and claims. B. MI has decided the large-scale process reengineering effort is worth it so Annette can have more customer-facing opportunities. C. Annette's potential for human fallibility, and therefore MI's operational risk, is heightened by the effects of RPA systems. D. Annette still does most of these same tasks as MI has discovered she does them less expensively.
A
Before advances in machine learning and predictive analytics, homeowners policy underwriters were Select one: A. Limited in their ability to incorporate the many attributes needed to accurately predict losses. B. Able to incorporate more data into underwriting decisions and improve loss ratios. C. Finding new correlations between granular data about their customers and loss exposures. D. Helping insurers easily make a profit selling homeowners insurance.
A
Claims representatives can often find evidence that someone may be lying by Select one: A. Comparing his or her social media posts with his or her statements in a claim. B. Reading published lists of claims-fraud indicators. C. Controlling underwriting at the point of sale. D. Increasing the claims staff.
A
Combining certain artificial intelligence technologies with robotic process automation to replicate human judgment best broadly describes which one of the following? Select one: A. Intelligent process management B. Natural language processing C. Computer vision D. Business process management
A
Establishing a sense of urgency helps managers Select one: A. Persuade key individuals to invest in the change and creates the momentum required to spur the organization to action. B. Pay closer attention to the concerns expressed by employees about customer-facing issues and how they can be resolved. C. When purchasing hardware and software for data analytics projects because the new equipment will encourage employee buy-in. D. Stick to the organization's priorities and develop solutions that include and engage most members of the staff.
A
In detecting claims fraud with data analysis, an insurer should be prepared to Select one: A. Reevaluate the attributes in a predictive model. B. Raise rates to cover fraudulent claims. C. Identify all claims in a cluster as fraudulent. D. Increase its claims staff.
A
In seeking to identify claims that will potentially develop into complex and more costly claims, the data science team at Great Midsouthwest Insurance ranks a series of attributes by their importance to the data model. Which one of the following terms describes the level of predictive power of each attribute? Select one: A. Information gain B. Lift C. Statistical relevance D. Precision factor
A
In the model training process, which one of the following is used to test the model and determine its predictive power on data that was not used in model development? Select one: A. Holdout data B. Information gain C. Supervised learning D. Training data
A
Millstone Insurance (MI) wants to identify potential ways to improve its claims handling process. Process mining begins with data analysis of claims activity logs to identify the claims handling activities and the timing of them. Which one of the following would help MI set the initial benchmarks for each activity in the current claims handling process? Select one: A. Process map B. Process tree C. Cluster analysis D. Classification tree
A
Park Slope Baking recently began manufacturing a line of gluten-free products. Caroline is the products liability underwriter for the account and is not sure how concerned she should be about the new product line. She knows that gluten-free products have become more popular recently, but she does not know if there have been any products liability problems. Which one of the following is an appropriate technique that data scientists can use to help Caroline discover emerging risks with gluten-free products? Select one: A. Cluster analysis B. Classification tree C. Telematics D. Predictive modeling
A
Personal lines insurers are investigating ways to incorporate discounts for homeowners insurance policies based on homeowners' use of which one of the following? Select one: A. Smart home devices B. Loss ratios C. Classification analysis D. Predictive modeling
A
Ralph's Creative Pools is trying to assess any potential online damage following a water pollution scare in the city where it is headquartered. It has hired a vendor to search various articles to conduct text mining of social media and various articles and perform a sentiment analysis of the various mentions. Through this effort, Ralph learns that the significance of a certain term found in an abundance of mentions is referred to as Select one: A. Inverse document frequency. B. Closeness. C. Term frequency. D. Corpus.
A
Wearables rely on which one of the following types emerging technology to respond in a specific way when a threshold of detection or measurement is reached? Select one: A. Smart sensors B. Radio frequency identification C. Lidar D. Computer vision
A
Which one of the following is an algorithm used to group data into clusters of claims? Select one: A. K-means B. Unsupervised learning C. Network analysis D. Centroid
A
Which one of the following is particularly useful in vacant or temporarily unoccupied buildings? Select one: A. Temperature sensors B. Motion sensors C. Power consumption sensors D. Light sensors
A
Which one of the following smart products can be used by a claims adjuster to gather data about property damage and geographic locations that are difficult to reach? Select one: A. Drone B. Wearable C. Smart phone D. Robot
A
A traditional analysis technique that, when paired with criticality analysis, helps an organization prevent or reduce the severity of a harmful event by identifying the order in which critical failures should be addressed is Select one: A. Root cause analysis. B. Failure mode and effects analysis. C. System safety analysis. D. Fault tree analysis.
B
A large manufacturing plant containing a variety of chemicals suffered a significant fire. Which one of the following smart products should be used to scan and inspect the building for structural damage and other unsafe conditions? Select one: A. Smartphone B. Robot C. Wrist monitor D. Accelerometer
B
A strengths, weaknesses, opportunities, and threats (SWOT) analysis focused specifically on the potential use of data analytics enables management to identify Select one: A. External strengths and weaknesses that could potentially be affected by a data analytics initiative. B. A "go" or "no go" conclusion for a specific project recommendation. C. Internal opportunities and threats that could potentially be affected by a data analytics initiative. D. Strengths and opportunities that can be paired to find ways to avoid or mitigate risks.
B
An insurer wants to understand what is causing adverse development of losses. Because the insurer doesn't understand the factors, what type of data analysis technique should it apply? Select one: A. Ultimate loss development B. Unsupervised learning C. Adverse development D. Supervised learning
B
An insurer's data analytics initiative should be guided by a vision statement that Select one: A. Directly improves the organization's products, services, cost controls, and relationship with its internal stakeholders. B. Succinctly aligns, directs, and inspires collective action toward a common end point that improves the organization's measurable results. C. Is desirable, feasible, and intuitive and alludes to appropriate plans and programs. D. Focuses on training, team creation, and other activities that produce positive organizational results.
B
By applying the enterprise risk management (ERM) process to a project, an organization Select one: A. Ensures that it will meet the needs of all the organization's internal stakeholders in developing the project. B. Synchronizes its overall risk management with the strategic goals and operational objectives of that specific project. C. Ensures that it will meet the needs of all the organization's external stakeholders in developing the project. D. Can assume that all possible current and future risks associated with the project have been analyzed and treated.
B
Durham Insurance wants to analyze and improve its auto claims handling process through process mining. Which one of the following would be an example of process instance that would be used in the analysis? Select one: A. Cancelled policy B. Reviewed policy C. Claimant name D. Insured name
B
If a manufacturer knows that a product could adversely affect a large number of people, it Select one: A. Has a defense that could reduce any damages in a product liability suit. B. Has a duty to provide an adequate warning on the product label. C. Can assign the product to the appropriate cluster. D. Has determined the negative sentiment score for the product.
B
Members of the center of excellence (COE) should have the power to Select one: A. Cancel the project if the organization does not have the current resources to execute it properly. B. Eliminate any obstacles to progress, expertise relevant to the change, credibility to convince others of the need for change, and leadership to catalyze the change process. C. Create a sense of urgency, even if it is artificial or exaggerated, to persuade all employees to invest in the change and develop momentum to spur the organization to action. D. Terminate any employees who do not embrace the change and integrate it into their work processes.
B
Upon completion of the data analytics project, Select one: A. Formal policies and organizational structures must be anchored in the past so staff will recognize how change has occurred. B. Individuals may not be able to see the broader perspective or effect on the change while it is taking place, so management may need to reinforce the vision. C. The organization's culture--the shared attitude established over the years that directs norms--should be preserved and articulated throughout the organization. D. New approaches will appear as disruptors to the status quo, and employees will be reluctant to embrace them.
B
Westfolk Mutual (WM) is a large personal lines insurer in New York. WM's management just learned of a news article that addresses the use of aftermarket auto parts in the state. The article discusses both the pros and cons of aftermarket auto parts. The article also names WM as one of many insurers in the state that encourage repair shops to use aftermarket parts whenever possible. Senior management at WM asks the risk manager to assess the potential reputation risk from the article. The risk manager works with data scientists to mine news articles and social media for any mention of WM's association with aftermarket parts over the last 24 hours. Which one of the following types of analysis will the risk manager attempt to do once the data is collected? Select one: A. Root cause analysis B. Sentiment analysis C. Cluster analysis D. Regression analysis
B
Which one of the following is a typical result of an insurer's implementing a flat rate increase for all of its homeowners customers? Select one: A. Customers with lower loss ratios will have a hard time finding coverage with other insurers. B. Customers with lower loss ratios will find less expensive coverage elsewhere. C. The overall loss ratio for the book of business will improve. D. Customers with higher loss ratios will stay, increasing profitability.
B
Which one of the following statements is true regarding smart operations? Select one: A. Smart operations are very gradually making their way into the field of risk management. B. Applying algorithms to big data generates the information underlying smart operations. C. The preventive analytics that lie at the heart of smart operations are based on historical data. D. Smart operations are limited by the amount of data that they can gather and process.
B
Which one of the following statements regarding process mining is true? Select one: A. Process mining is the same as typical data mining in that both apply results to a specific area. B. Process mining differs from typical data mining in that process mining applies results to an entire process. C. Process mining is applied to claims to verify whether the settlement amount was correct. D. Process mining cannot usually be applied effectively to the claims handling process.
B
Which one of the following types of workers compensation claims are the most difficult for insurers to identify at the time of first report? Select one: A. Catastrophic claims B. Potentially complex claims C. Medical-only claims D. Potentially minor claims
B
A data analytics technique that uses supervised learning and defined target variables to segment data according to known attributes and determine the probability of a workplace accident is Select one: A. A root cause analysis. B. A fault tree analysis. C. A classification tree. D. An information gain.
C
After data from sensors is categorized to develop attributes that help predict workplace accidents, these attributes are combined with traditional workplace accident attributes and then analyzed independently for their Select one: A. Target variables. B. Holdout data. C. Information gain. D. Footsteps shift.
C
As the data analytics initiative progresses, the project manager compares the quality, time, budget, and other constraints established for the project with the project status and determines whether the project goal will be achieved or whether resources should be reallocated to achieve the goal. These activities are part of Select one: A. Scanning the internal and external environments to determine how the data analytics project will affect stakeholders. B. Identifying the risks associated with the data analytics project. C. Monitoring and reviewing the data analytics project. D. Analyzing the risk associated with the data analytics project to determine the source, cause, and consequences of each risk.
C
Cluster analysis is Select one: A. A type of supervised learning. B. A predictive model. C. An unsupervised learning technique. D. A way of obtaining information through language recognition.
C
Often, machine learning Select one: A. Is not appropriate for smaller insurers, which do not have access to large amounts of data. B. Implements a flat rate increase for all of an insurer's homeowners policies. C. Reveals relationship and trends not found through traditional ratemaking analysis. D. Is simply a ratio that measures loss and loss adjustment expenses against earned premiums.
C
Telematics devices track driving habits and then Select one: A. Classify each insured auto for its principal use. B. Perform sophisticated modeling techniques to make that information relevant. C. Transmit that information wirelessly to a computer. D. Divide auto insurance applicants into homogeneous classes.
C
The goal of business process management (BPM) is to Select one: A. Achieve short-term objectives, such as closing claims. B. Issue decisions regarding individual claims. C. Optimize a business process, such as claims handling. D. Change the major activities in the claims handling process.
C
The leadership team for a major transformation in an organization is usually made up of Select one: A. The top-level managers for each department and for each team within the organization. B. All management-level staff who expect to have time available to devote to convincing all staff of the importance of the project. C. One highly visible change agent and a collaborative team of sponsors of the change. D. Those individuals who are known to be disruptors within the organization.
C
Three Hills Insurance (THI) is a regional insurer. The profitability of its homeowners book has been declining for several years. In the past, THI has been able to maintain profitability by using traditional rating variables and implementing flat rate increases across the entire book when necessary. It has been suggested that THI use machine learning to segment its homeowners book. Based on its relatively small size, management does not believe that it has sufficient data available to benefit from machine learning. Which one of the following is a valuable source of data, not used in traditional methods, that could be made available to THI through machine learning? A. Account loss ratios B. Customer credit ratings C. Producers' and claim representatives' files D. Policyholder applications
C
Time estimates for crucial activities in the data analytics project should be assigned acceptable variances to be referenced in the event that they could cause results to fall outside the acceptable parameters. These estimates are part of Select one: A. Monitoring the progression of the data analytics project to assure that the goals will be achieved. B. Identifying the risks associated with the data analytics project. C. Analyzing the risks associated with the data analytics project. D. Scanning the external environment for a sense of how the data analytics project will affect the organization's stakeholders.
C
To gain a better understanding of what is involved in the investigation and resolution of different types of claims, the insurer will need to Select one: A. Use predictive modeling to determine claims outcomes. B. Create a process map showing the results of process discovery. C. Combine process mining with data analysis techniques D. Determine the average amount of time for each process instance.
C
To hone a data analytics project's goal, the risk manager could Select one: A. Examine any legal and regulatory issues associated with employees' and contractors' personal data and the mining of social networking websites. B. Interview the organization's leadership team to better understand the desirable outcome of the project. C. Interview various internal stakeholders to develop a series of quantifiable objectives that, together, constitute the project's overall goal. D. Examine any technological issues that must be resolved to make the project feasible.
C
Traditional fraud indicators Select one: A. Are represented by lines in a network. B. Analyze social network connections. C. Depend on fraud that has happened in the past. D. Are proof of fraud.
C
Upper management and the center of excellence (COE) should eliminate barriers to change, such as Select one: A. An organizational culture that recognizes the necessary shift from the status quo and supports retirement of traditional methods of leadership. B. The assignment of roles to formal and informal leaders within the organization to lead and implement a data analytics project and to monitor it after implementation. C. Technological entanglements that limit change, like legacy computers and communications systems that discourage interaction among departments. D. Goals for implementing data analytics techniques that stretch the capabilities of employees and persuade them to explore the new techniques.
C
Vehicle telematics can function as Select one: A. Generalized linear models. B. A way for customers to maintain privacy. C. Ongoing driver education for those who participate. D. Standard class loss ratios.
C
Which one of the following emerging technologies has led to major advances in the field of automotive safety and the development of automated vehicles? Select one: A. Artificial intelligence B. Wireless sensor networks C. Computer vision D. Drones
C
Which one of the following explains why a computer recursively applies a model? Select one: A. To identify attributes that can be used B. To analyze claims data from previous years C. To analyze different splits in the values of attributes D. To determine the probability of a target variable
C
Which one of the following is an example of a wearable? Select one: A. Heart monitors B. Unmanned aircraft C. Helmets that monitor fatigue D. Robots
C
Which one of the following is artificial intelligence in which computers continually teach themselves to make better decisions based on previous results and new data? Select one: A. Usage-based insurance B. Telematics C. Machine learning D. Generalized linear model
C
Which one of the following uses infrared light to detect nearby objects? Select one: A. Drones B. Wearables C. Lidar D. Robots
C
A generalized linear model is Select one: A. A type of clustering. B. A type of unsupervised learning. C. Used to calculate total incurred values. D. A type of supervised learning.
D
Competitors' pricing, an increase in expected hurricane-related claims, and a shrinking talent pool for industry hiring would be included in which one of these SWOT factors? Select one: A. Opportunities B. Weaknesses C. Strengths D. Threats
D
For which one of the following is it difficult to rely solely on traditional underwriting guidelines? Select one: A. Any products that are made, sold, distributed, or given away B. Products with warning labels C. Products that do not have uncertain risk characteristics D. Newly introduced products and emerging technologies
D
In developing a predictive model for complex claims, the data science team at first uses only a portion of the claims data. This is called Select one: A. Holdout data. B. Provisional data. C. Trial data D. Training data.
D
In scanning the environment, the risk manager for an organization's data analytics initiative first needs to develop Select one: A. A strategy to inform affected employees about the initiative and encourage their participation in the project, while recruiting data scientists and support personnel. B. A list of specific categories of data that the project intends to analyze and the means of acquiring the necessary data. C. An idea of the number of risks that are associated with the initiative and of those which will require treatment. D. A sense of how the initiative will merge with its existing infrastructure and how the project could affect the organization's stakeholders.
D
Management and the center of excellence (COE) may use milestones in a data analytics project to Select one: A. Represent stretch goals that employees can work toward; by reaching them, employees achieve more than should be expected. B. Scale down the project to smaller goals that work units can accomplish and avoid overstressing employees by communicating the longer-term, overarching project goal. C. Designate levels of data that can be mined effectively to extract crucial information that fulfills the insurer's needs. D. Represent visible improvements or short-term accomplishments that are part of the progress toward the project goal.
D
Predictions of ultimate losses for specific accident years are based on Select one: A. Paid losses. B. Loss adjustment expenses. C. Adverse development. D. Incurred losses.
D
Reputation is Select one: A. The risk that negative publicity, whether true or not, will damage a company. B. Independent of the trust and confidence of key stakeholders. C. A perception that does not change over time. D. An organization's social license to, or not to, operate.
D
Risk avoidance, modification, transfer, retention, or exploitation, as well as contingency planning, are Select one: A. Some of the factors that contribute to the external environment of the data analytics project. B. Component activities associated with the data analytics project and are used to analyze the risk. C. Methods used to monitor progression of the entire project to assure that the project goal will be achieved. D. Some of the major options for treating risk and keeping the entire project on track within variance requirements.
D
Which one of the following explains how classification tree analysis can be used to improve the claims handling process? Select one: A. Classification tree analysis is an unsupervised learning technique that can explore unknown attributes. B. Classification tree analysis can be used to determine the probability of outcomes for new claims. C. Classification tree analysis uses the process map to recommend improvements for the overall workflow. D. Classification tree analysis can be used to assign new claims according to target variables.
D
Which one of the following is an approach used to search social media for references to products? Select one: A. Predictive modeling B. K-means C. Underwriting guidelines D. Text mining
D
Which one of the following is an attribute that could help an insurer better understand homeowners wildfire claims? Select one: A. Claims history B. Year built C. Age of plumbing system D. Relative wind speed
D
Which one of the following is correct regarding k-means clustering? Select one: A. K-means clustering is a type of supervised learning. B. K-means clustering is a type of generalized linear model. C. K-means clustering is used when the factors are understood. D. K-means clustering is used when there is not a target variable.
D
Which one of the following is the measurement of how often a term appears in a document? Select one: A. Betweeness B. Inverse document C. Sentiment analysis D. Term frequency
D
Which one of the following is the primary rating factor for personal auto insurance that is determined by where an auto is normally used and garaged? Select one: A. Radius of operation B. Use of the auto C. Special industry classification D. Territory
D
Which one of the following is the study of the nodes and edges in a network? Select one: A. A centroid B. A fraud ring C. Unsupervised learning D. Network analysis
D
Wyatt Insurance sells workers compensation policies to small regional manufacturing companies and recently began gathering and analyzing its underwriting and claims data using artificial intelligence (AI). Which one of the following would be an appropriate AI application for Wyatt going forward? Select one: A. Wyatt uses chatbots and conversational AI to assess photos of claimants and their injuries to reduce the time needed for providing a proper diagnosis. B. With assisted intelligence, systems act on their own to complete tasks and make decisions without human involvement. C. Wyatt uses automated intelligence to identify casualty exposures and provide risk-rating recommendations, as it learns based on underwriters' decisions. D. Chatbots and conversational AI can be used to engage in conversations with Wyatt's customers and claimants to gain information or answers to questions.
D
