Google Search Appliance software version 6.0
Posted June 2009
The Google Search Appliance enables you to provide universal search to your users. You can get the most from your Google Search Appliance by using some or all of its many features to fine-tune and enhance universal search. Become familiar with the Google Search Appliance's features by reading this document and apply those features that best suit your search solution.
A user's search experience is based on accessing the Google Search Appliance to enter a search and receive results. The Google Search Appliance provides many built-in features that ensure a satisfactory search experience for users. For a list of these features, refer to Built-In Search Appliance Features.
However, the Google Search Appliance comes with many features that you can use to enhance the user's search experience, especially the way in which the search appliance returns results. Ways that you can enhance the search experience include:
The framework that you use for enhancing the search experience is the front end.
After you deploy one or more search experiences, you can use the advanced search reporting feature to gather feedback about how users are searching, You can use this feedback to refine and enhance the search experience.
This section briefly describes each feature that you can use to enhance the search experience and contains links that you can follow to get more information about each feature.
Without any administrator intervention, the Google Search Appliance provides a rich search experience by using its built-in search features. The following table lists these built-in search features.
| Feature | Description |
|---|---|
| Automatic spell check | The search appliance automatically suggests spelling corrections accurately, even on company-specific words and phrases. The spell checker can suggest corrections in multiple languages, including U.S. English, French, Italian, German, Spanish, Portuguese, and Dutch. |
| Sorting search results based on relevance | The search appliance finds the highest quality and most relevant documents for a search query; Google factors in more than 100 variables for each query. |
| Automatic filtering of duplicate snippets | If multiple documents contain identical titles, as well as the same information in their snippets, only the most relevant document of that set is displayed in the results. |
| Automatic filtering of duplicate directories | If there are many results in a single web directory, then only the two most relevant results for the directory are displayed. An output flag indicates that more results are available from that directory. |
| Automatic filtering of languages | Limits search to a specified language, as determined by the majority language used in the web document body. |
| Dynamic page summaries | Users can judge relevance of results more easily with dynamically generated snippets showing a query in the context of the page. |
| Results grouping- | Users can navigate search results easily and clearly using intelligent grouping of documents residing in the same narrow subdirectories. |
| Cached pages | Users can view search results even when the sites are down by using cached copies of pages included in the search results. |
| Highlighted query terms | User can quickly find the most relevant section of a document by using the highlighted query terms displayed on cached documents. |
| View as HTML | Users can display documents without needing the original client application of the file format because the search appliance automatically converts over 220 file formats into HTML. |
| Sort by date | User can access time-sensitive information first by using date sorting. |
| Advanced Search page | Users can perform complex and sophisticated queries with over 10 special query terms, including Boolean AND, OR, and NOT searches. |
The simplest way to give access to the search features of the Google Search Appliance is to use the default search and results pages. These pages provide a user interface that is simple and intuitive and most users are familiar with them from their experience with Google.com. For more information about these default pages, refer to Customizing the User Interface.
The URL for the default search page is http://SEARCH_APPLIANCE_NAME. For example, if your search appliance is named compgsa, the built-in search page is available at http://compgsa. You can also link this URL from your website to provide user access to it.
A front end is a framework that manages most of the elements of a single search experience, including:
In each front end, you can use different search appliance features to create a unique search experience for a specific user group. For example, suppose that you want to create different front ends for users in Sales and users in Engineering, based on their different requirements.
Users in Sales need to be able to search the entire index, but would like certain types of documents (presentations, forecasts, and product documentation) to appear at the top of the search results. Users in Engineering need to search only a subset of the index and would like technical documents to appear at the top of the search results.
To ensure that each user group sees preferred documents at the top of the search results, you might use various search appliance features, including KeyMatches and result biasing. To segment the index so that users in Engineering only search the part of the index that interests them, use a search appliance feature called "collections."
Most of the search appliance features described in this section are associated with a front end, including features that give users feedback on their searches and features that refine search results. You create and manage search experiences by using anywhere from a few to all front end features.
There is no limit to the number of front ends that a single search appliance can have. You can create multiple front ends to deploy multiple search experiences for users.
Set up a front end by performing the following steps with the Admin Console:
For in-depth information about setting up and using front ends, refer to Managing the Search Experience in Creating the Search Experience: Introduction.
The Google Search Appliance enables you to force certain documents to the top of search results with its KeyMatch feature. When users search with a term that you specify, the search appliance always presents the KeyMatch first. Users can navigate immediately to the featured document and spend less time searching and viewing less relevant documents.
For example, suppose you administer a search appliance for Altostrat, a fictional company. This business is a reseller of the Google Search Appliance. Internally, when the technical team searches on the term "gsa," they are most interested in navigating to http://pm.altostrat.com/products/gsa or https://pm.altostrat.com/products/products. To help users find these URLs, you can provide the links as KeyMatches, as shown in the following figure.
You also might consider using KeyMatches to promote documents that are too new to be in the search index or might not appear among the highest search results.
Because a KeyMatch is specific to a front end, you can aim a KeyMatch at a specific group of users.
You set up a KeyMatch by matching a search term to a specific URL and specifying a title for the match. In the preceding example, there are two KeyMatches for the search term "gsa." In the first KeyMatch:
In the second KeyMatch:
There can be up to five KeyMatches for a single search term, and you can associate up to five URLs for each KeyMatch. However, a maximum of three KeyMatches are returned for a search.
Set up a KeyMatch by performing the following steps with the Admin Console:
For in-depth information about setting up and using KeyMatches, refer to Using KeyMatches to Guide Users to URLs in Creating the Search Experience: Best Practices.
The Google Search Appliance can suggest alternative search terms to users for their original keyword searches with its related queries feature. When users search with that term, the search appliance always presents the related query at the top of the search results. When a user clicks the related query, the search appliance runs the search again and returns additional results.
For example, suppose some of Altostrat's content contains official names, such as "Google Search Appliance," as well as informal terms, such as "gsa," for the same products. Users might use either official product names or informal terms when they perform searches. To help users find all relevant documents and not miss any, you can provide a related query, as shown in the following figure.
You also might consider using related queries to suggest your own product names and company-specific acronyms or terminology.
Because a related query is specific to a front end, you can aim a related query at a specific group of users.
Set up a related query by performing the following steps with the Admin Console:
For in-depth information about setting up and using related queries, refer to Using Related Queries to Suggest Alternative Searches in Creating the Search Experience: Best Practices.
The Google Search Appliance can group search results by topic with its dynamic result clusters feature. By clicking a result within a group, a user can:
For example, suppose a user who looking for information about the expense budget for NYSSA. He searches for this information using the term " expense nyssa." A dynamic result cluster appears with the results, as shown in the following figure.
By default, dynamic result clusters are disabled. For any front end, you can:
If you have implemented KeyMatches or OneBox modules, you might want to place the dynamic result clusters to the right of search results. This placement minimizes user scrolling down the page for natural search results.
Set up dynamic result clusters by performing the following steps with the Admin Console:
For in-depth information about setting up and using dynamic result clusters, refer to Using Dynamic Result Clusters to Narrow Searches in Creating the Search Experience: Best Practices.
In some instances, the most relevant result for a search query is real-time, structured data, such as delivery tracking numbers. This type of data does not usually reside in the search index because it would be obsolete before it could be indexed. Using Google OneBox for Enterprise, you can provide users with access to real time business data from various systems.
A OneBox module sends the user’s query to a different collection on the same appliance or from an external source, such as an application or database, and retrieves relevant data immediately. A OneBox module is returned when a user's search term matches a "trigger" term.
OneBox results appear above other results on a user's results page, formatted according to an XSLT stylesheet template. The following figure shows the OneBox that displays on Google.com when a user searches for american 102.
Google enables you to define a OneBox module that appears for certain search terms or the OneBox can appear for every search query. A OneBox module can either search a collection or access a URL for a site that returns XML results.
The Google Enterprise Developer Community has developed a number of publicly available OneBox modules that you can download. To view a list of these OneBox modules, visit the OneBox gallery on code.google.com.
Before you can use a OneBox module with an external data source, a developer must create a OneBox provider, a program that receives queries from the appliance, obtains data from the application, and returns results to the appliance. The OneBox module can call an internal or external provider:
Before you can set up a OneBox module, you must choose a front end where you want to implement it. For information about front ends, refer to Using Front Ends.
Set up a OneBox module by:
Having a clear definition of what you are trying to achieve with a OneBox module is essential to a successful implementation. The Google OneBox for Enterprise Design Principles provides information you can use to design a OneBox module.
For information about creating internal or external providers, refer to OneBox for Enterprise Developer's Guide.
The Google Search Appliance enables you to restrict search results by:
Filters help ensure that the search appliance serves appropriate results to users.
For example, you might want to restrict search results by domain to ensure that searches in various regions return only results with local information. Suppose you want to restrict results on the pages in the United Kingdom to show only products and special offers available there, so you create a front end for users in the U.K. Suppose the domain name for the U.K. is www.mycompany.com.uk. You might use this domain name to create a domain filter so that when users in the U.K. perform a search, only results that match the domain name appear. All domains ending with the name are filtered.
Because a filter is specific to a front end, you can aim a filter at a specific group of users.
Set up a filter by performing the following steps with the Admin Console:
For in-depth information about setting up and using filters, refer to Using Filters to Restrict Search Results in Creating the Search Experience: Best Practices.
The Google Search Appliance can automatically widen a search by adding terms that are synonymous with the search terms through query expansion. Query expansion helps users get search results that they would otherwise miss.
When a user searches on a term, the search appliance expands the search to include synonymous terms. For example, a user searches on the term "documentation," and the search appliance returns the most relevant results that contain the keyword "documentation." However, the user misses results that contain alternative terms, such as "guide" and "manual." If the search term "documentation" is expanded to include "guide," "guides," "manual," and "manuals," the search is wider and returns an increased number of relevant results.
Google dictionaries of synonyms for English, French, Italian, German, Spanish, and Portuguese are built into the search appliance. Whenever a user enters a search query that matches a synonym in one of these languages, the term is expanded.
However, you can create and upload custom synonym lists to improve search quality. You can also create and upload a blacklist file, which contains terms that should not be expanded.
After you have configured and enabled the appropriate query expansion files, you can set a query expansion policy for a front end. A query expansion policy determines the synonym or blacklist files that are used with a front end.
Set up query expansion by performing the following steps with the Admin Console:
For in-depth information about setting up Query Expansion and uploading custom synonym lists, refer to Using Query Expansion to Widen Searches in Creating the Search Experience: Best Practices.
The Google Search Appliance enables you to influence the order in which the Google Search Appliance ranks search results with its result biasing feature. Result biasing helps ensure that users see results that are relevant to their interests or roles.
For example, given the search term "AltoStrat," you might want code or design documents to appear high in search results for the engineering group, while you might want product specifications to appear higher for the marketing group. The following figure illustrates two different search results rankings for the same search term, "AltoStrat."
Because result biasing is specific to a front end, you can aim result biasing at a specific group of users.
The Google Search Appliance provides three ways for you to influence results ranking:
Because increasing the relevancy of specific documents may also decrease the relevancy of others, use result biasing to boost scores of content sources that you are certain are more authoritative than other sources.
Set up result biasing by performing the following steps with the Admin Console:
For in-depth information about setting up and using result biasing, refer to Using Result Biasing to Influence Result Ranking in Creating the Search Experience: Best Practices.
The Google Search Appliance enables you to divide your search index into sections and provide search across different content to different groups of users with its collections feature. A collection is a subset of the complete search index.
Search results from a collection have the same relevance ranking as full index searches. Only the content searched differs because it is restricted to the individual collection's content. By searching a collection, users get relevant search results more quickly than by searching the entire index.
Using collections, you can show different results to different users. For example, you might create collections such as:
To search a collection, a user can select the collection name from a pull-down menu on the search box, as illustrated in the following figure.
Collections are independent of front ends. However, you can use a custom front end with a specific collection to help improve searches and enhance results.
To set up a collection, specify URL patterns that it should include. If you set up more than one collection, the same entry can appear in multiple collections. You can set up an unlimited number of collections for a search index.
Set up a collection by performing the following steps with the Admin Console:
Before you add a pull-down menu for searching by collection, you must choose a front end where you want to implement it. Add a collection menu by performing the following steps:
For in-depth information about setting up and using collections, refer to the following topics in Google Search Appliance documentation:
The Google Search Appliance enables users to monitor topics of interest with its alerts feature. Alerts are email updates of the latest relevant search results based on a user's topic of interest.
A user sets up an alert by clicking My Alerts on the search page, logging in to the search appliance by using her LDAP user name and password, and choosing an hourly, weekly, or monthly schedule. After the user creates an alert, the search appliance sends the user an email whenever it finds new or changed documents about the topic of interest.
Alerts require that the user authenticate using their LDAP credentials. This means that an LDAP server, populated with the users email must be present and accessible by the search appliance for Alerts to function.
Set up alerts by performing the following steps with the Admin Console:
For an in-depth information about setting up and using alerts, refer to Providing Alerts for End Users in Creating the Search Experience: Best Practices.
The Google Search Appliance offers a default user interface that is simple and intuitive, like Google.com. The following figure illustrates the default search and results pages.
A Google Search Appliance user interface is associated with a single front end. The search appliance has a default front end, which uses the default user interface. You can use the default user interface without any customization. However, a search appliance can have multiple front ends, each with its own, customized user interface.
For example, you can customize a front end by making changes that reflect your organization's visual identity, such as by using your logo and your color scheme. Other types of changes that you can make to the user interface include:
The AltoStrat examples in this document present a customized user interface, as shown in the following figure.
After the search appliance receives and executes a search query:
An XSLT stylesheet contains information about which elements should appear in the user interface and how the elements should look. Each front end can use the same stylesheet or a different stylesheet. Each search appliance front end has a default XSLT stylesheet, which can be used with any front end.
You can customize a user interface by editing an XSLT stylesheet with one of the following tools:
Both of these tools are accessible from the Serving > Front Ends > Output Format page in the Admin Console.
Changes that you make using the Page Layout Helper are fully supported by Google Enterprise Support. If you want to contact support about changes made using the Page Layout Helper, file a help ticket. Use can also refer issues to the Google Search Appliance group on Google Groups. Changes that you make using the XSLT Stylesheet Editor are not supported by Google Enterprise Support. If you have issues about changes made using the XSLT Stylesheet Editor, you can refer them to the appropriate Google Group.
Even if you do not have any special knowledge of XSLT, you can effectively customize a Google Search Appliance user interface using the Page Layout Helper. Use the Page Layout Helper to perform the following tasks:
The Page Layout Helper has a Preview button that you can click to open a browser window and see how the page will look when you save your changes.
The following figure shows the Global Attributes section of the Page Layout Helper.
Take note that the Page Layout Helper enables you to change only some of the elements in the XSLT stylesheet. If you want to make extensive changes to a search appliance user interface, you need to work directly in the XSLT Stylesheet.
However, you can start by customizing the user interface using the Page Layout Helper. After you finish making and saving changes in the Page Layout Helper, you can make further changes in the XSLT Stylesheet Editor. Any changes that you make with the Page Layout Helper appear in the XSLT stylesheet.
If the elements that you want to change are not available in the Page Layout Helper, you must use the XSLT Stylesheet Editor to change them. This editor enables you to make changes directly in the XSLT stylesheet. The XSLT stylesheet contains sections for various components, preceded by comments so that you know whether a section can be customized.
To work in the XSLT Stylesheet editor, you need knowledge of XSLT, XML, and HTML.
The following figure shows the XSLT Stylesheet Editor.
Edit an XSLT stylesheet by performing the following steps with the Admin Console:
For in-depth information about using the Page Layout Help, the XSLT Stylesheet Editor, and making changes to the user interface, refer to Creating the Search Experience: Customizing the User Interface.
The Google Search Appliance's advanced search reporting feature enables you to gather information about user clicks on search results. By using advanced search reporting, you can determine:
By analyzing user clicks, you can also identify ways to improve the search experience.
For example, suppose information about user clicks shows that a users in range of IP addresses are all searching for information about project with the code name of "Antilles." None of the users are finding a satisfactory URL at the top of the search results. Some of the users are finding a satisfactory URL on page 3 of the results. However, many users are abandoning the search after viewing page 2 of the results.
The range of IP addresses tells you that the users who are searching for the results are all Engineers in the U.S. who are working on project Antilles. You might create a front end for this group of users and force results about project Antilles to the top of the results using KeyMatches.
Once you enable advanced search reporting, the search appliance modifies search result pages by inserting code that tracks all links that a user clicks. From that point on, when a user clicks a link in the search results, those selections are tracked for analysis later on.
You can export an advanced search report. Each entry in an advanced search report represents a single user click or other action, such as page load, in the search appliance user interface.
The reports provide rich search behavior information such as:
Set up advanced search reporting by performing the following steps with the Admin Console:
For in-depth information about setting up and using advanced search reporting, refer to the following Google Search Appliance documentation:
Let Google know what you think about this document by sending feedback to gsadoc-gtm-feedback@google.com.
|
