Plan a new Funnelback for higher education search

This guide applies to the Squiz Experience Cloud version of Funnelback for higher education. See the v15.x Funnelback for higher education documentation for earlier releases.

Creating a new Funnelback for higher education search requires a bit of forward planning that must be carried out before attempting to setup a new search.

The planning must determine what higher education content sources are available and the data for these will be accessed for inclusion in the search. Without this information it is not possible to use the setup tool to effectively create your search as you need to know all of the data sources and their types when you define them with the tool.

In addition to this it is worthwhile spending time to determine what metadata is available, how these will map to the pre-defined metadata classes and what (if any) additional metadata classes should be defined.

Components

Funnelback for higher education is divided into multiple components:

  • A set of content sources (e.g. courses, events, faculty), with predefined metadata class mappings.

  • A set of reusable display component templates (e.g. tabs, facets, video results, course results)

  • A results page that defines the search results template and static resources.

  • A search package that draws together the different content sources which are included in the search.

Step 1 - Figure out which data sources can be included

Review the list of content sources below. For each of these think about whether or not the university has a suitable content source and if it is appropriate to include it in the search.

These are the ones included in Funnelback for higher education:

Websites

One or more websites that are crawled using a web crawler.

Courses

Course or program information. Structured data is recommended.

Events

Events information. Structured data is recommended.

People

People profiles. Structured data is recommended.

Facebook

Facebook pages to include in the search. Requires Facebook API tokens.

Instagram

Instagram feeds to include in the search. Requires Instagram API tokens.

Twitter

Twitter feeds to include in the search. Requires Twitter API tokens.

YouTube

YouTube channels to include in the search. Requires YouTube API tokens.

Structured data means that fielded data is available for the content source. This might be web pages that have metadata fields indicating values such as first and last names or might be JSON or XML data that contains key value pairs of fielded information. The fielded information is used to provide rich information in the search results and also for structured auto-completion. Data sources that lack this data will have reduced functionality.

Step 2 - Examine the content sources

For each of these content sources:

  • Determine how the content source will be accessed, and the appropriate data source type. e.g. For a people database will the data be indexed by crawling a set of web pages containing staff profiles (web data source), by directly accessing an XML export from a web accessible link from the database (web data source) or by an API (custom data source)? This information is required when you create your higher education search as you need to indicate what type of data source to create.

  • Review the default metadata classes that are defined for the content type and determine which ones are available in the content source. Also note what the source metadata fields are as these will need to be mapped to the pre-defined metadata classes for the data source.

  • If you are including a web content source as well as course, people or events content sources consider if any of these will be captured when including the web content, and if it is appropriate to exclude courses, people or events web pages from the web content that is indexed to avoid potential duplicate content within the search index.

© 2015- Squiz Pty Ltd