Download Relativity Analytics Specialist Exam and more Exams Nursing in PDF only on Docsity!
Relativity Analytics Specialist Exam When you first access the Active Learning queue, your set of 200 documents will appear in the list. True/False. False - only coded documents appear in the the document list view. What would prevent a document in the Active Learning queue from being assigned to an active reviewer? (Select All That Apply) a. Document has already been coded b. Document is assigned to a logged off user c. Document is assigned to another user d. User does not have appropriate permissions a. Document has already been coded c. Document is assigned to another user d. User does not have appropriate permissions Active Learning works best when you analyze the content of family members when making a Responsive or Not Responsive call. True/False.
False. Reviewers should apply the "four corners" rule when reviewing documents in an Active Learning project. If you wanted to use family-based review (Prioritized Review Only), then entire families must be added to the saved search of culled documents used for the Analytics Classification Index. True/False. True, otherwise the family members will not be included. Reviewers can change the coding decision on documents they previously reviewed in an Active Learning project. True/False. True - The next model build will include the most recent coding update. Active learning model only builds once you have at least five documents coded with the positive choice and five with the negative choice. True/False. True, this is the minimum for an active learning model to build. You can turn off family-based review after Prioritized review has started. True/False.
For a standard request for production with a need to do redaction and/or privilege review, which type of Active learning workflow should you choose, coverage or prioritized review? Prioritized Review. If you have a document set with lower richness, then should you chose prioritized review or coverage review? Prioritized Review Coverage review serves up both relevant and non- relevant documents. True/False. False - coverage review serves up the documents the model is most unsure about. Which of the following projects should you use Coverage review for? (Select All That Apply) a. Classify an Incoming Production b. Responding to a Second Request c. Project with Low Richness d. Project requiring Family Based review a. Classify an Incoming Production b. Responding to a Second Request
Coverage review works best with document sets with lower richness? True/False. False - Use prioritized review For LARGE Active Learning projects, which of the following could make updating ranks and model rebuilding faster? (Select All That Apply) a. Cull documents and create sub-project d. Take a Random Sample c. Delete Prior Ranks Yourself d. Suppress Duplicates a. Cull documents and create sub-project d. Take a Random Sample c. Delete Prior Ranks Yourself The buffer of documents in queue for reviewers is based on the number of active reviewers. True/False. False - it is based on the number of total assigned reviewers At the start of an active learning project, you can estimate richness by coding a sample set of documents and dividing the total # of documents by the # of
c. 80 percent d. 20 percent b. 10 percent Prioritized Review serves up what percentage of highest-ranking un-coded documents? a. 70 percent b. 10 percent c. 80 percent d. 20 percent a. 70 percent Order the following steps for adding or removing documents from a prioritized review project? a. Navigate to the Analytics Tab b. Run, then Incremental c. Click the Classification Index used to create the Active Learning Project d. Pause Review d. Pause Review a. Navigate to the Analytics Tab c. Click the Classification Index used to create the
Active Learning Project b. Run, then Incremental Turning off the health index documents will raise the percentage of highly ranked documents served to 100% from 70%. True/False. True - it would remove the documents that would have been served to your reviewers for the purpose of improving predictions (the 10% random documents and 20% with mid-range scores). You can use Active Learning to find any outliers that might exist in previously coded data by searching for documents coded in the reviewer field as Responsive and Classified by the system as Not Responsive. True/False. True - you can set an active learning project to QC previously coded data When you update the responsive cutoff value, the value is updated in which of the following places where it's used in the application? (Select All that Apply) a. Project Settings b. Project Home Dashboard
40 and 60 until the model begins training. As the model learns throughout the project life cycle, the Rank Distribution chart is expected to gravitate toward 0 or 100 depending on how documents are coded on the positive choice or negative choice. Typically, a well trained model will have a large pile of documents centered around ________, and another larger pile centered around _______. (Select All that Apply) a. 90 or 95 b. 70 or 75 c. 20 or 25 d. 30 or 35 b. 70 or 75 c. 20 or 25 In a Document Rank distribution chart, the Y axis gives what Active learning statistic? a. Document Distribution b. Designation c. Relevance Rank d. Coding Decision c. Relevance Rank
The Prioritized Review Progress chart tracks the ability of the model to find relevant documents over time. It does this by monitoring the reviewers' coding decisions on the high-ranking documents chosen by the Prioritized Review queue. As a general trend, you should expect to see the relevance rate decline over the course of the review. True/False. True - However, you may see a spike in the relevance rate if a large amount of new documents are added to the project, or if the definition of relevance changes during the course of the review. Documents included in the Active Learning queue for index health are excluded from the relevance rate calculation. True/False. True The Prioritized Review Chart only updates when documents are coded in the Prioritized Review Queue. True/False. True - if only manually-selected documents are coded, the Prioritized Review Progress chart won't display relevance rate data.
For which of the following Projects does it make sense for the Inactive Queue Retraining setting to be turned on at an interval of 24 hours? a. Projects with routine coding outside the queues b. Projects with occasional coding outside the queues c. Projects with no other coding activity d. Projects with active coding within the queues a. Projects with routine coding outside the queues b. Projects with occasional coding outside the queues For which of the following Projects does it make sense for the Inactive Queue Retraining setting to be turned off? a. Projects with routine coding outside the queues b. Projects with occasional coding outside the queues c. Projects with no other coding activity d. Projects with active coding within the queues c. Projects with no other coding activity d. Projects with active coding within the queues Which of the following events trigger a model rebuild? (Select All That Apply)
a. Coding in an Active Queue b. Turning on Inactive Queue Retraining c. Starting project validation d. No Coding Activity a. Coding in an Active Queue b. Turning on Inactive Queue Retraining c. Starting project validation Rebuilding the model occurs on the back end and updates the relevance ranks of documents. True/False. False - Rebuilding the model occurs on the back end, and it recalculates the relevance scores that control which documents are served up in the queue. For a Machine Classification Against Coding Values Widget, you should sort of which of the following fields for the stacked bar chart display? a. Active Learning Designation b. Grand Total c. CSR - Active Learning Project Cat. Set::Category Rank d. Active Learning Project Cat. Set::Category Rank (Ascending) d. Active Learning Project Cat. Set::Category Rank (Ascending)
False - Update Only New Documents (Incremental Build) is enabled by default but it can be disabled to include all the documents in the build. When automating the build of a conceptual index, which other Automated Workflow should you run prior to Running an Analytics Index? a. dtSearch Index b. Classification Index c. Structured Analytics Set d. Search Term Reports c. Structured Analytics Set Running a build on a classification index will cancel any in-progress active learning project validation. True/False. True - this should be kept in mind when setting up an automated classification index. Enabling the Hide Images for QC option prevents users without sufficient permissions from viewing the document images in the selected Imaging Set. True/False.
True - this can be selected when creating an Automated Workflow for running an Image Set. How is the Estimated Total Relevant Documents? The estimated number of relevant documents is calculated by projecting the richness rate across the whole project. What is the engine behind Active Learning? a. Relativity Algorithm b. Support Vector Machine c. Classification Index d. Search Terms Report b. Support Vector Machine When running project validation, what does the Elusion Rate statistic measure? a. The percentage of relevant documents across the whole sample. b. The percentage of truly positive documents which were found by the active learning process. c. The percentage of documents coded relevant in the uncoded, low-ranking portion of the sample.
c. Group By: Active Learning Designation d. Display: Stacked Bar Chart b. Pivot On: Active Learning Designation d. Display: Stacked Bar Chart You can use the Machine Classification against Coding Values widget to determine when a project is done by looking for few or no purple bars (Not Reviewed) over the 50 mark for relevance. True/False. True - under the 50 mark having been Machine classified as likely not relevant. If you estimated richness at the start of your project, you have a rough estimate of how many relevant documents are in your project. The number of documents coded relevant should be close to the richness estimate when the project is done. True/False. True - as Richness is the percentage of relevant documents across the whole sample. Which of the following search conditions when combined with the AND operator will produce documents that have been skipped by reviewers during Active Learning review? a. AL Designation is Not Set
b. Active Learning Project Reviewers > These Conditions > Active Learning Project Reviewers is Not Set c. Active Learning Project Reviewers > These Conditions
Active Learning Project Reviewers is Set d. Active Learning Completed is Set a. AL Designation is Not Set c. Active Learning Project Reviewers > These Conditions Active Learning Project Reviewers is Set What is the Discard pile? The set of uncoded documents with ranks below the rank cutoff when Project Validation is started. What is the Discard Pile size? The number of documents below the rank cutoff that are not coded when Project Validation was started. What is the Discard-pile elusion rate? The percentage of documents in the discard pile that are relevant. What is the sample elusion rate? The percentage of documents in the elusion sample that are relevant.
