ACTUAL Exam Questions and CORRECT
Answers
2 set of Docs required for AI - CORRECT ANSWER - 1. Training Set
2. Searchable Set
5 steps in analytics lifecycle - CORRECT ANSWER - 1. data organization
2. building an index
3. data analysis
4. doc review
5. doc qc
Training Set - CORRECT ANSWER - Doc set that index learns from. Determine concepts
of index. Foundation for searchable set.
Searchable Set - CORRECT ANSWER - On documents you want analytics to function
Concepts in doc determine position of searchable set.
5 Video Stats to Buidl AI - CORRECT ANSWER - 1. Create Training
2. Create Searchable
3. Create AP
4. Apply Filters
5. Pop & Build Index
5 Analytic Operations - CORRECT ANSWER - 1. Analytics categorization sets
2. Clustering
3. Concept searching
,4. Keyword expansion
5. Similar document detention
9 Index Creation fields - CORRECT ANSWER - 1. Name
2. Order
3. Analytics Profile
4. RA Server
5. Email Notification Recipients
6. Training Set
7. Searchable Set
8. Optimize Training Set
9. Auto remove English email signatures & footers
Searchable set default - CORRECT ANSWER - default ________returns only extracted
text field
When must you apply filters in AI? - CORRECT ANSWER - Apply __________ in the
analytics profile b4 you hit full or incremental indexing
Optimize Training Set - CORRECT ANSWER - Auto includes only conceptually valuable
docs gets rid of conceptually irrelevant training docs
(4) Automatically remove English email signatures and footers - CORRECT ANSWER -
(1) Enables email header filter
(2) Disables Go Words Filter
(3) Disables OCR filter
(4) Excludes docs greater than 30MB
, 2 different defaults for remove English email signatures and footers - CORRECT
ANSWER - 1. Yes for new
2. No for old
(3) Training Set Settings - CORRECT ANSWER - (1) Exclude docs over 2MB of
extracted text. Use script Set Extracted Text Size Field
(2) Only authored content.
(3) If not authored content or conceptual data then useless. AVOID Zip, Excel
10 Non-authored data to avoid in Training Set - CORRECT ANSWER - (a) Compressed
ZIP/RAR
(b) System files
(c) Excel files mostly numbers
(d) Image files
(e) CAD drawings
(f) Maps
(g) Calendar items
(h) Docs w/ OCR errors
(i) Poor quality OCR or handwritten
(j) Docs w/ little text below (0.2 KB)
Docs Optimization Excludes - CORRECT ANSWER - o Very short/long docs
o List w/ bunch of numbers
o Spreadsheet
o System log files
o Text resulting from processing errors
4 things that Optimize training look for? - CORRECT ANSWER - 1. word count
2. uniqueness