The Query function enables you to search the dataset for
specific words, concepts, and phrases. You
can use several powerful query techniques, including Word Query, Query
by Example (QBE), and Query in Notes.
To
open the Query window, perform the following procedure.
Click the Show Query
button on the IN-SPIRE main window toolbar, or from the
Tools menu on the IN-SPIRE main window, select Query.
The Query window will display with the Word Query tab visible. The
main query types display in tabs across the top of the Query window.
Click
on a tab to display the related query tool.
Word Query looks for documents that contain
combinations of Query Text words. Queries can be constructed using the
Boolean operators AND,OR,and NOT. If no Boolean operator is specified,
query words are joined with the Boolean AND, resulting in each document
in the result set containing all the query words. Word
Query finds exact matches only.
Query By Example enables you to find documents
having related content rather than exact word matches. To do this, it
creates a "mathematical signature" of the Query terms and then
looks for documents having a similar signature. The Query Results slider
controls the number of hits and how similar the result set documents will
be.
Query
Words in Notes enables you to query the notes that may be attached to
documents in your dataset. You
can query on phrases, use wildcard characters such as *, and construct
a proximity query using the ~ character.
All query text is interpreted in a case-insensitive manner, which means
capitalization is ignored. This
applies to all types of queries, including inside phrases.
Constructing
a Word Query
A Word Query enables you to search for words
in a document. To
create a Word Query, perform the following procedure.
If
necessary, click on the Word Query tab. The
Word Query tool will display.
Type
a word or words in the Query Text box.
Select the Fields you want to search.
You can
either search All Fields (the default) or select one or more fields to
search by clicking on the items in the Fields list using Ctrl-click. The
terms are combined with the AND operator and all selected fields are searched.
The Query
button will be enabled.
Select whether you want to Query all
documents or Query only selected documents by clicking on the option under
the Query Text box.
Click Query. The
query will run, with the results being highlighted in the Galaxy.
In addition to steps 1-4, you can use
any of the following refining techniques to perform a Word query.
From the Summary word list (see Summary)
select one or more terms and click Word Query.
Type a phrase in quotes, for example, "islamic
jihad", in the Query Text box, and select one or more fields to search.
Within the phrase you may use an asterisk (*) as a wildcard.
Connect words in the Query Text box with AND,
OR, or AND NOT, to construct a Boolean query (! is ignored, as are XOR,
and |).
Use an asterisk at the end of a word for word
stemming. For example, "defen*" will return documents containing
any words beginning with "defen", including "defense",
"defend", "defensive", etc. Note: Since IN-SPIRE removes punctuation, contractions
such as "don't" are seen as 2 words, "don" and "t".
Use wildcards to find possible contractions.
Reuse one of the queries in the Query History
list. Click on the query you want to use; it appears in the Query Text
box.
Reuse a saved Query Set.
Click Group Results to create a group in
the Groups tool composed of the documents
in your query results.
Word Queries and Fields
If you pick more than one field, it is important to note that the entire
query is run against each of the specified fields separately. Such
a query returns documents for which one or more of the selected fields
satisfies the query. The
query is not applied across the specified fields. For
example, suppose you selected the TITLE and AUTHOR fields, and then entered
the following query:
Abbott
and Costello
This would return documents with both words, Abbot and Costello, in
the title, or both words in the author field, or both words in both fields.
Note that
this query would not return a document with the following field content:
TITLE:
The Life
and Times of Bud Abbott
AUTHOR:
I. B. Costello
This is because neither field, taken by itself, satisfies the query.
One effect
of this rule is that selecting all of the individual fields in the Fields
list is not equivalent to selecting All Fields.
Understanding Relevance in Word Queries
Word Query can quantify and provide relevance feedback from Boolean
queries, analogous to and consistent with the "closeness” quantity
computed for Query by Example (QBE) results. Relevance
is made up of the following components:
Independence: The
relevance of each document returned by a query is independent of any other
document returned by that query.
Coverage: The
more query terms found in a document, the higher its relevance.
Balance: An
even distribution of query terms found in a document outranks an uneven
distribution.
Complexity: The
more complex the query, the less coverage required for a document to achieve
a given relevance score.
Size: Shorter
documents will have higher relevance scores, all other things being equal.
Customizing Punctuation Rules
Punctuation rules apply not only to the query expression, but to the
content of any document to be searched.
Phrase queries take full advantage of the power and flexibility of the
punctuation rules, without sacrificing
accuracy.
For example, suppose you want to “de-hyphenate” all hyphenated words,
such that “hold-up” maps to “holdup.” You
can customize the “middle” rule for the hyphen such that it is null rather
than a space. Now,
a document that contains both "holdup” and “hold-up” would respond
exactly the same to a query for the word holdup, the word hold-up, the
phrase "holdup” “hold-up.”
Constructing a Query
by Example
Query by Example enables you to find documents similar to or "close
to" a document you have by searching the dataset for documents containing
words or phrases similar to an example you supply. To
create a Query By Example, perform
the following procedure.
Click
the By Example tab. The
Query by Example panel will display. Text
highlighted in blue indicates major terms.
In the Galaxy, select the document or documents
you want to use as an example.
Open the Document Viewer.
In the Document Viewer text area, find a phrase
you think significant. Highlight
the phrase you want to use as your query, and right-click. A
drop-down menu will display.
Select Query by Example from the drop-down menu.
The Query will run and the results will display
in the Query by Example tab in the Query window.
The initial results will pinpoint the one document in the dataset that
supplied the phrase on which you based the query.
To see documents that contain words and phrases
similar to the ones you selected as the query, move the Query Results
slider to the right. The
number of documents containing similar words will go up as you move the
slider and the tally is displayed under the Quantity axes of the Query
Results histogram.
The documents will highlight in the Galaxy as you move the slider.
Note that
as the Quantity of documents goes up, the Quality of results will go down.
Click the Group Results button to form a group
in the Groups tool of the documents you
found with the Query by Example.
View the documents in the new group with the Document
Viewer.
Querying Within Document
Notes
Query in Notes enables you to find words and phrases in notes attached
to documents. These
are notes you or colleagues may have created. If
a dataset has not been annotated, a query on the notes will yield no results.
To create
a Query in Notes, perform the following
procedure.
Click the
Notes tab. The
Notes panel will display.
Enter the word or phrase you want to search for
in the appended document notes.
Select whether you want to Query all documents
or Query only selected documents by clicking on the option under the Query
Text box.
Click Query. The
query will run, with the results being highlighted in the Galaxy.
Click Group Results to create a group in
the Groups tool composed of the documents
in your query results.
Performing Typical
Query Activities
The Query tool offers a
number of tools that enable you to refine your document searches.
To create a group
with the results of a query, click the Group Results button. The new
group will display as a locked group in the Groups window under the Selection
History folder. The
new group will have the same name as the query.
Queries are automatically saved in the Query
History list as you create them.
Using an Asterisk in a
Query
You can use an asterisk in a phrase query as a wild card placeholder
as a substitute for either characters or words. Use an asterisk between
words and followed by a number, to take the place of a number of (unspecified)
words. For example:
Query
Example Results
"word1 *4
word2" will return documents containing from 0 to 4 unspecified
words between word1 and word2.
For example,
"New York *4 team"
the New York team is...
New York Rangers
team members
New York Yankees
baseball team
New York Mets
famous baseball team
New York State
civil service employees' team
Since IN-SPIRE removes punctuation, contractions such as "don't"
are seen as 2 words, "don" and "t". Use
wildcards such as *2 as described in the above table to find possible
contractions.
Working
with the Query History Window
The Query History panel of the Queries window keeps a list of all the
queries you run on the dataset. The
Query History will display the type of query and the number of document
'hits' the query yielded.
W indicates a Word query
Q indicates a Query by Example
N indicates a Query within Notes
F indicates a query on a Field such as Title,
rather than All Fields.
You can rerun a query at any time by selecting it from the list and
clicking the Query button.
You can select a Word query and run it as a Query in Notes or as a Query
by Example to yield different views into the dataset.
Deleting a Query from
the Query History
In the Query History list of queries you have
run, click on the query you want to delete.
Click Remove The
query will be removed from the Query History list.
Translating Queries
If you have a translation engine installed on your system or server,
you can use the integrated translation capabilities
of IN-SPIRE to formulate bilingual queries. Once
you have the translation
turned on, perform the following procedure.
Load a foreign language dataset.
Once translation is turned on and you have selected
the language you want to translate, click the Query button.
The Translate button in the Query window becomes
active.
Enter a Query in the Query Text box. For
this example, we will perform a Word Query on a Spanish language dataset.
To query
the dataset for the word 'horse', we typed the term in the Query Text
field.
Click the Translate button. The
query terms will be translated and will display in both languages.
Click the Query button. The
translated query will be run on the document dataset and the results will
display in the Query History.
Click Group Results to group the documents found
by the bilingual query. You
can view these documents in both languages in the Document
Viewer.
on the IN-SPIRE main window toolbar, or from the
Tools menu on the IN-SPIRE main window, select Query.
Note
