|
|
PART ONE: Welcome to TheLaw.net and Search Basics
If you've never conducted electronic searching, we think you will find
it easy to learn and surprisingly efficient. If you have prior
experience with Westlaw® or LEXIS®,
that experience will be helpful, but remember, TheLaw.net is a different
service. Even though each of these services can provide you with court
cases, each has a unique way of doing it. That will affect your strategy
and search construction.
Electronic Searching Strategy
Searching is a process, not an event.
This should be your mantra when using TheLaw.net. Searching a library is
not about spending time and mental energy formulating the "golden
query" that retrieves your desired information in a single stroke.
In practice, good online searching involves formulating a succession of
queries until you are satisfied with the results. As you view results
from one search, you will come across additional leads that you did not
identify in your original search. You can incorporate these new terms
into your existing query or create a new one. After each query, evaluate
its success by asking:
- Did I find what I was looking for?
- What better information could still be out there?
- How can I refine my query to find better information?
Issuing multiple queries can be frustrating or rewarding, depending on
how long it takes you to identify the key material you need to answer
your research problem.
Full-Text Libraries and Documents Defined
A full-text library is a collection of related whole documents
assembled into a single searchable unit. The individual documents can be
massive or minuscule, but they should bear some relation to each other
(e.g., court opinions issued from the same jurisdiction). A full-text
library is composed of smaller units called documents. When you search a
database, you will retrieve documents that contain information that
matches your query request.
Virtual Library: Searching Multiple Libraries
With TheLaw.net, your query can search in multiple libraries open at
once; this is known as a virtual library. With a virtual library
searching will be slightly slower than searching in a single library. In
addition, documents retrieved from different jurisdictions are combined
into one Hit List.
Indexing
TheLaw.net, Lexis and Westlaw all depend on the word as an essential
tool to search and retrieve documents. And, like Lexis and Westlaw,
TheLaw.net uses the principle of word indexing. During the publication
process, an indexer goes through every document and creates an index of
every word in every document and also tabulates how many times each word
is used in each document. When you do a search, you are not really
searching through the full-text of the documents; you are searching the
word index of the documents.
Stop Words
As opposed to a keyword-based system, TheLaw.net uses a full-text
retrieval software, meaning that it indexes every word in a document
with the exception of Stop words. Stop words are those terms that are
programmed to be ignored during the indexing and retrieval processes, in
order to prevent the retrieval of extraneous documents. Generally, a
stop word list includes articles, pronouns, adjectives, adverbs and
prepositions ("the", "they", "very",
"not", "of", etc.) that are most common words in the
English language. Using stop words in full-text searching is vital in
the context of Relevance Ranking, as described below.
Relevance Ranking
The most powerful weapon in the searcher's arsenal is Relevance Ranking.
Simply put, relevance ranking lists a set of retrieved documents so that
the documents most likely to be relevant are shown to you first.
Remember, Relevance Ranking is not an indication of legal relevance.
Relevance Ranking arranges documents based on the mathematical
measurement of similarity between your query and the content of each
record. What determines the likelihood of relevance? An analysis of the
database is performed using a combination of the following indicators:
- Breadth of Match - Documents containing more of the various query
terms are weighted more relevant.
- Inverse Document Frequency - Documents containing terms which
occur less frequently in the entire database are weighted more
relevant.
- Frequency - Documents with a higher occurrence of a query term are
weighted more relevant.
- Density - The comparable length of retrieved documents is
calculated to apply a higher relevancy weight.
In this analysis, stop words are ignored. This reduces the time spent
processing your search and prevents an artificial boost of relevance to
what are actually irrelevant documents, since "the" would
probably retrieve every record in a database.
The researcher receives several benefits from Relevance Ranking. With
the assistance of Relevance
Ranking, you will find the most relevant documents in the shortest
period of time. And, as you read down the Hit List, once you determine
the documents are getting less applicable, you can stop reading results
of this search because you know you have already viewed the most
relevant documents. Finally…you don't have to be a computer expert who
can compose the most complex of queries in order to find valuable
information! Next |
|
Database Manual
Search Syntax
Field Searching
Browser Basics
Power User Tips
Drop Down Menus