The flexibility offered by the search criteria in Discovery Attender allows users to fine tune criteria to pinpoint specific documents that satisfy eDiscovery and other requirements. In this latest installment of the series, highlighting the lesser known features of Discovery Attender (see previous posts for other topics), we will explore the choices available using the post-search filtering options that helps users further refine their result sets. The features dedicated to filtering are found in the Result Management screen. The different views – by keyword, location, extension, unique items etc. – provide distinct sub-set snapshots of results. This post, however, will focus on the standard and advanced filters.
Most list views in the Result Management screen include standard filters. This functionality is accessed by selecting the Filter menu item from the tool bar or context (Right-Click) menu.
There are two basic types of standard filters. The first, ‘By Selected Location’ will limit the view to those results that match the location of the highlighted item. This is useful if you have searched multiple data stores but need to isolate one. Two clicks and the filter is applied.
The second type of standard filter is ‘By Mark’. To understand the workings of this filter type, it is useful to know that a Mark is a symbol defined by the user (in the Settings) that can be associated with individual results. The Marks can then be used as the basis for a new filter to show any results that match the chosen Marks, or those that match no marks at all.
Since the user chooses the meaning of the icons, there are many opportunities for organization and filtering with the Mark functionality. A number of helpful features in Discovery Attender utilize the Mark options including Comparing Searches and Automatic Labeling.
Aside from these standard offerings, culling results can also be performed with the Advanced Filters. These filters are created via a wizard. They are used to selectively cull the result set using criteria based metadata such as address, document properties and dates which are stored in the back-end database as the main Discovery Attender search completes. If the results were indexed (a choice in the search wizard), then results can also be refined by keywords.
Selecting the Advanced Filter node in the navigation pane brings up a view where filters can be created, copied or deleted using the toolbar or context menu. Creating a new filter opens a wizard screen which guides users through the criteria choices.
Although the Advanced Result Filter may seem like a stripped down version of the main Discovery Attender search, the syntax and search methodology are different. Dates, addresses and other properties are resolved from the metadata stored in the databases, not the original document in the source data store. Logic for the Advanced Filter is also different than the standard search. This is especially true for searching of keywords which uses dtSearch index based syntax.
After an Advanced Result Filter is created, the results are displayed in a detail node created underneath the main Advanced Filter’s node. This sub-node is shown when a new filter has been applied, or, when you double click on a previously created filter. Used in conjunction with the Mark functionality, these filters provide flexibility for all result views.
Advanced Filters have many uses. Let’s say you just want to see a subset of your data within a certain date range. Since no one standard result view shows all the dates, the Advanced Filter becomes a one stop source to see related items. The same is true for addresses or domains. The Domain view is great for showing all results that match a particular domain, but not so helpful if multiple domains have the same keywords (e.g. yahoo.com, yahoo.co.uk, etc) The Advanced Filter can be used to combine all associated domains together in one view.
Using filtering in Discovery Attender is just one way to organize your data. Don’t hesitate to contact technical support with any questions, comments or if you’d like, more ideas on handling result sets.