Optimizing Processing

With content defined, the next thing to consider is the mechanics of how a query and the documents it returns will be processed. You may want to restrict access to documents or portions of documents, provide expansions to the query that a user enters, or record data to understand how the platform is being utilized. Some processing is under the user's control and some is internal to Watson™ Explorer Engine, but work together to determine the quality and scope of the results that are returned to the user.

Tip: Deciding on the type of processing that is required can be an evolving process, using reporting to understand how the platform is utilized will help you plan the future growth of your deployment.

The following table is designed to help you decide which features will be beneficial to you when optimizing processing. The left column of the table identifies a result that you may want to achieve, and the right column describes the feature(s) that can be utilized to achieve that goal.

Table 1. Processing Optimization Features

If you want to...

Then the feature you need to learn about is:

...restrict access to a document based on the original security of the document

Authentication - Watson Explorer Engine supports document and content level authentication with many different configuration options available.

...check for spelling errors and update the query based on the corrected words

Spelling Dictionaries - Queries can easily be mistyped. Configuring one or more spelling dictionaries to either manually or automatically correct these queries can significantly improve the user experience. Dictionaries can be created from using an existing word list and by using words/phrases collected from search collections in your platform.

...let users query with incomplete words using wildcards and/or regex

Wildcard Dictionaries - You can create wildcard dictionaries that take your user's queries and expand them based on matching the wildcard query to the dictionary. For example, a wildcard dictionary that contains hat, hot, and hut would produce a query of hat OR hot OR hut when the user specifies a query of h?t. Regular expression expansion can also be configured when querying Watson Explorer Engine search collections.

...add the querying of similar words and concepts to expand the scope of a search

The Terminology Manager module for Results Module - Terminology Manager lets Results Module users define relationships between words that can be use to automatically or manually expand user queries. The easy to use interface of Results Module makes it easy for any authorized user to update the semantic expansion of terms.

...block documents from being returned based on the content of the document

Knowledge Base of Stop Words - By creating a knowledge base of stop words (words or phrases that will cause an action to be performed when found in a result), you can compare the contents of any result against that list of stop words in order to prevent matching results from being returned to the user.

...allow users to connect to external search engines with the results provided in a single interface

Federated Search - You can connect the Watson Explorer Engine Platform Application to external search engines by creating sources that retrieve results from those sources, combine them, and present them through the unified Watson Explorer Engine interface.

...store a list of possible queries that can be suggested to users based on what they have entered into the search box...

Autocomplete - Watson Explorer Engine lets you collect phrases from search collections and word lists to create a dictionary of suggestions that can be offered when you enter a query. These suggestions can contain metadata such as images, descriptions, and even direct links to the suggested document.

...return specific search results or other content based on the user's query

Key Match Sources - Watson Explorer Engine sources are highly customizable, one option is to create sources that match a specific key. When that key is triggered, the source is used to deliver content to the user. That content can be results from a collection, federated results, or even static information held within the source.

...allow users to specify syntax to limit a query to specific contents

Syntax - Defining syntax lets the user enter queries that are restricted to certain fields or certain types of queries.

...collect historical and live data on usage and system events

Reporting - You can collect many forms of historical data by using the Watson Explorer Engine reporting tools. These reports can be scheduled using the Watson Explorer Engine scheduler and automatically delivered in multiple formats.