Download RELATIVITY ANALYTICS SPECIALIST EXAM LATEST and more Exams Advanced Education in PDF only on Docsity!
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
What can you do to get a more stable and good model more quickly? Turn Suppress duplicates on Disadvantage of suppressing duplicates Need to review suppressed docs using another method. Use prioritzed review when...
- Quickly locate and review the most relevant docs
- Family based review
- Doc set with low richness
- unsure which queue to select Coverage review works best with docs greather richness then... >2% How are documents served with prioritized review? 10% random 20% from the middle range (40 - 60) 70% highest ranking To run prioritised review wfam 3 considerations?
- Must be a new queue
- After start review edit the settings is not possible
- Families with more than 100 docs shouldnt be included
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
SHould you use other relational groups during prioritized review then group identifier while reviewing family? NO Should you tag only the relevant document relevant while doing prioritized review with family? Yes, dont tag the whole family How to add new docs when review is already started for prioritized and coverage?
- Pause review
- Add or remove
- Run incremental on the analytics index.
- Start review How are documents served with coverage review? 45% docs above 50 45% docs below 50 10% docs exact 50 How will serve coverage review docs with the below and above the rank of 50? It will serve the docs closest to 50. To make coverage review efficient do 2 things.
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
- Richness - > If you have estimated richness at the start of the project you have enough estimate of how many relevant docs are in you project. De coded relevant docs should be close to the richness estimate when the project is done. Index health he number of index health documents reviewed in the Prioritized Review queue. These documents are excluded from the relevance rate calculation. Highest Ranked the number of highly ranked documents reviewed in the Prioritized Review queue. Highest Ranked Coded [Positive Choice] the number of highly ranked documents that were coded with the positive designation in the Prioritized Review queue. Relevance Rate the percentage of documents that were chosen for being highly ranked that were then confirmed as relevant by reviewers' coding decisions. Estimated Eluded Documents (Range)
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
This estimates the number of relevant documents you would miss if you produced all documents marked relevant, as well as those with ranks at or above the cutoff. Precision Rate (Range) the percentage of found documents which are truly positive Recall Rate (Range at CL80%) the percentage of truly positive documents which were found by the Active Learning process Precision Margin of Error (CL80%) the margin of error for precision as estimated from the sample size, the equivalent portion of the whole project, and the observed precision rate on the validation sample. Richness Rate (Range) the percentage of documents which are relevant (positive choice). Richness Margin of Error (CL95%) the margin of error for richness as estimated from the sample size, the whole project size, and the observed richness rate on the sample Model updates Sections contains the history of active learning model
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
Richness measures... the overall relevance rate of all documents in the project. As an Active Learning project progresses, these relevant documents will be found in different buckets, but the overall percentage of relevant documents remains the same. Recall measures.... the percentage of truly positive documents which were found by the Active Learning process. Precision measures.... the accuracy of what you're planning to produce. Two sample validations? Fixed & statistical Why conceptual analytics It helps you organize and asess the semantic content of large, diverse and/or unknown sets of docs. Difference between Conceptual & Structured
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
Structured relies on the specific structure of the content while conceptual focuses on the related concepts within the docs. Even if the dont share the same key terms and phrases. Two types of analtycis indexes? Conceptual and classification Do analytics indexes need an dtSearch? No Conceptual analytics? Uses Latent Semantic Indexing to discover concepts between documents. This indexing process is based solely on term co-occurrence. The language, concepts, and relationships are defined entirely by the contents of your documents and learned by the index. Classification index uses coded examples to build a Support Vector Machine (SVM) to predict a document's relevance. This index is used solely by the Active Learning application. Classification indexes learn how terms are related to categories based on the contents of your documents and coding decisions made within the Active Learning project Do Conceptual Analytics require a Analytics index? Yes
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
200 max 1000 When checked Optimize Training Set... It will automatically includes only conceptually valuable docs in the training source while excluding conceptually irrelevant docs. Why could you use Optimize training set It prevents you from manually remove irrelevant documents Dimensions in analytics index Determines the dimensions of the concept space into which documents are mapped when the index is built. More dimensions increase the .... and refine the .... Conceptual values & Relationships Default setting for Dimension 100 What happens when you increase more dimensions? It can lead to more nuances due to more subtle correlations that the system detects between docs. Does removing English email signatures and footers impact the speed?
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
Yes, enabling this option increases the population speed. Is it recommend to turn off Remove english email sig and footers? No, highly not recommend. Enable email header filter? Removes email header fields Does enable email header filter removes the subject line? No Is the Enable email header filter recommended to turn off? No If you want to disable enable email header filter you must... Set remove english email sig and footers to no STop words Determines words you dont want in your coneptual index Saved search conditions to train the data source for conceptual or classificaiton
- Only include Extracted text as the field. IF OCR or translation is present add that as well.
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
Full Does incremantal force Analytics to go through every stage? No Important things to consider prior incremental? Quantity (don't add to many docs) and Subject matter (only related docs otherwise train the model again. What is a repeated content filter finds the text wherever it occurs in each document that matches your configuration parameters and suppresses it from the Analytics index Why would you use repeated content filters To include only authored content which isnt overshadowed by content such as confidentially footers or standard texts. If repeated content isnt included it can ... Create false relationships between documents A regex filter in repeated content will Removes content matching the pattern. Will repeated content filters be applied to SAS> No
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
Can repeated content filters be applied to dtsearch No Clustering is used to... Create groups of conceptually similair docs. With clustering you can identifiy... Conceptual groups in a workspace Compared to analytics categorization sets clustering doenst require? Example docs or category definitions How are the clustered groups formed? conceptual similarity Clustering is usefull when working with... unfamiliar datasets How is the cluster group called with docs outside the cluster? Unclustered Does clustering culls down irrelevant data? No Clustering_Title Format?
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
Documents are tightly conceptually-related that higher level clusters. Default and max value for hierarchy depth Default: 3 & Max: Clustering_minimum coherence? the level of conceptual correlation that items must have to be included in the same cluster Default value for clustering coherence?
What does the analytics engine do to a cluster when the minimum coherence is below the value? It breaks up the cluster into subclusters ONLY if the max depth has not been reached What does the analytics engine do to a cluster when the minimum coherence is on or above? It leaves it allow. Generality determines how vague or specific the cluster will be at each level Genarality values from?
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
0 to 1 Default value Genarality
Cluster genarality value closer to 0.0 means... More clusteres that are tighter at each level of the cluster tree High cluster generality gives you... fewer and broader clusters A low generality means... Many top-level clusters are created A high generality means.. Few top level cluster are created 3 types of cluster visuals? Circle pack Dial visualization Nearby clusters You can use concept searching to... find information without precisly phrased query. Benefits of concept searching
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
to provide a lists of highly correlated terms, synonyms or strongly related terms in your document set. When you submit a block of text you get... a list of single terms that are strongly related to that content. You can use Analytics Categorization Sets... to create a set of example documents that analytics uses as the basis for identifying and grouping conceptually similar docs. Categorization is useful when... early in review project when you understand key concepts of a case and identify docs that are representative concepts. Can you designate examples and add them to various categories of Categorization sets? yes To how many categories can you designate an document? 5 Is Categorization effective if you have identified categories or issues of interest? Yes
END OF PAGE
ANALYTICS
SPECIALIST EXAM
LATEST
is categorization effective if you know how you want to title the categories Yes Is categorization effective if you have one or more focused example docs to represent the conceptual topic of each category? Yes Is categorization effective if you have one or more large sets of data that you want to categorize rapidly without any user input after setting up the category scheme. yes 3 example of docs for Categorization sets?
- Focused on a single concept
- Descriptive
- Free of distracting text Document example focused on a single concept? represent a single concept relevant to the category Descriptive document example? Fully represent the single concept it is defining. Focus on at least one to two fully developed paragraphs.
