Magnify Enhancements

In this section:

Magnify is an Enterprise Search solution that helps users find business content across the entire organization and then understand that information using Business Intelligence. Users are provided a single entry point to search structured and unstructured data stored in nearly any location. The Magnify search page is a simple and interactive user interface that dynamically categorizes search results with various navigation, sorting, and viewing options, such as analyzing search results in a tabular format with the added ability to chart and filter an array of metrics and dimensions. Users can access real-time information by drilling down on links to WebFOCUS reports.


Top of page

x
Sentiment Analysis Support

How to:

Sentiment Analysis determines the general tone (positive, negative, or neutral) of an indexed record using natural language processing. Businesses can use the results of a sentiment analysis to help track feedback on products or services that are often captured using social media tools, such as blogs, wikis, and forums.

Magnify integrates with the DataFacet Taxonomy Server for Sentiment (installed separately) to perform sentiment analysis on structured and unstructured content. As content is indexed by Magnify, the DataFacet Taxonomy Server is called using a web service to determine the overall expressed feeling of the content as positive, negative, or neutral. The DataFacet Taxonomy Server returns a score to Magnify, which is stored as a meta tag with each search result. The score of the content is displayed as part of the Magnify Category Tree. This allows users to filter search results based on the overall tone of the information. Additionally, you can sort the search results by the sentiment score.



x
Procedure: How to Set Up the Magnify Plug-in for Sentiment Analysis

As part of the FORMAT MAGNIFY procedure, select and assign one or more fields to be analyzed for sentiment. Content is then passed as part of the indexing process. The score is returned to the index and is used as part of the search functionality.


Top of page

x
Flat File Indexing and Synchronization

How to:

The Adapter for Flat File has been enhanced to monitor file repositories so that administrators can index file repositories using Magnify and keep them synchronized as new files are added, modified, renamed, moved, or deleted in the repository.



x
Procedure: How to Configure the Adapter for Flat File to Monitor File Repositories for Search

The following steps provide an overview of the required configuration steps:

  1. Create an application mapping to the file repository directory.
  2. Create and configure a Flat File synonym for file monitoring.
  3. Point the Flat File synonym to monitor the file repository created by application mapping in step 1.
  4. Configure a FORMAT MAGNIFY procedure that uses the Adapter for Flat File synonym to index the application mapping directory that is being monitored.

Top of page

x
Dynamic Partitioning Support

How to:

As the size of an index library increases, the search speed decreases and can eventually cause the indexing or search process to stop working. The dynamic partitioning feature enables the scalability of local index files by splitting index folders into partitions and thereby increasing the performance of the indexing and search processes significantly. For example, once a feed process starts for a data source called retail, new index folders named retail1, retail2, retail3, and so on, are created dynamically.



x
Procedure: How to Enable Dynamic Partitioning

The following steps provide an overview of the required configuration steps to enable dynamic partitioning:

  1. Set the data source value in the WebFOCUS report used for indexing or in the iWay Service Manager Service Object.
  2. Specify the index folder name in the collections.xml file.
  3. Using the Magnify Console, enable the Dynamic Partition feature, and specify an index size for each library.

Top of page

x
Massive Parallel Search Support

Massive parallel searching enables you to distribute Lucene indexes in order to search across many machines in a network. This feature increases the speed at which large amounts of data can be searched. It is best used in a distributed network environment, where indexes are fed independently on each remote server.

The following points summarize the setup of a typical massive parallel searching solution:


WebFOCUS