Chp5 Overview of Text Data Access Flashcards
The general goal of data access
The general goal of text data access is to connect users with the right information at the right time.
Data Access Approach
pull
push
pull and push
In pull mode, the user initiates the access process to find the relevant text data, typically by using a search engine. This mode of text access is essential when a user has an ad hoc information need, i.e., a temporary information need that might disappear once the need is satisfied.
In the push mode, the system initiates the process to recommend a set of relevant information items to the user. This mode of information access is generally more useful to satisfy a long-standing information need of a user or analyst.
Multimode Interactive Access
Ideally, the system should provide support for users to have multimode interactive access to relevant text data so that the push and pull modes are integrated in the same information access environment, and querying and browsing are also seamlessly integrated to provide maximum flexibility to users and allow them to query and browse at will.
text retrieval vs database retrieval
In databases, the data are structured where each field has a clearly defined meaning according to a schema. Thus, the data can be viewed as a table with well-specified columns.
In contrast, the data managed by a search engine are unstructured text which can be difficult for computers to understand.
A database query clearly specifies the constraints on the fields of the data table, and thus the expected retrieval results (answers to the query) are very well specified with no ambiguity. In a search engine, however, the queries are generally keyword queries, which are only a vague specification of what documents should be returned.
Finally, the results are different. In database search, we can retrieve very specific data elements; in TR, we are generally only able to retrieve a set of relevant documents.
