Report:Engineering Village/Viewing Results/The Hit List
From Intellogist
| Report | Patent Coverage Map | Ratings | Comments |
| This search system report was created by the Intellogist Team and is available for viewing only. If you'd like to share your knowledge on Intellogist, please visit the Best Practices, Glossary, or Community Reports pages. Registered users may be notified of any substantial changes to this report by placing a "watch" on the Revisions page, which is the last page listed in the table of contents. To learn more about using the Intellogist "watchlist," see the Watchlist Help page. | |
  |
|
The Hit List
By default, Engineering Village displays the result set 25 records at a time, with a ‘Results Manager’ box at the top of the list.
This section will discuss how this list of results can be manipulated and filtered to show the most relevant hits, and to visualize the trends that are taking place in the data set. The next section will discuss display options for records of particular interest in more detail.
Sorting by Relevance
The relevance sort uses an algorithm that takes into account the following:
- Whether the search terms are found together as an exact phrase or separately
- When terms are found separately, closer proximity ranks higher
- The number of times that the term/phrase appears in the record
- The term’s location within the record (terms found at the beginning of the field rank higher than terms found towards the end)
- Whether the terms are found within fields designated as particularly relevant, i.e., the title field
- How often the term appears in the database as a whole (terms used often are less relevant than terms that are less common)
Filtering and Visualizing Data Trends: Facets
Engineering Village offers a unique way to further refine a result set by the use of a feature called “Facets”. Facets enable the user to further refine a result set and optionally view charts or download accompanying data. They can be used to refine a result set by basic bibliographic categories including Database, Author, Author Affiliation, Year, and can also be used to filter data sets by special indexing that may be applied via database producers.
The figure below shows a facets panel, which appears at the far right hand of the results set. The functionality of this panel is simple: users review the top data points from each aspect of the results set (author, publication year, etc.) and then they are free to choose any of these to refine their results set.
The actions which can be taken from this bar are as follows:
- Choosing a data point and selecting “include” will filter the data set to include only those data points (for example, selecting the publication year 2008 and choosing include will produce a new data set made up only of the 741 documents published in 2008).
- Choosing a data point and selecting exclude will remove those records from the data set (for example, selecting the year 2008 and choosing “exclude” will remove those 741 documents published in the year 2008 from the data set).
- Entering a keyword term in the search bar at the bottom of the feature can perform an additional keyword search. Users may choose to filter their current results set by keyword, or to perform an entirely new search in “all content,” erasing their current data set.
In addition to these actions, the facets bar has some additional analysis and exporting functions. Specifically, the bar can be used to generate charts, showing statistical data about the results set in a graphical way. Any data on the Facets bar can be charted by selecting the small yellow chart icons, which appear next to each Facet data heading. The figure below shows a chart which was generated from author counts, via the Facets menu, on a keyword search for “Fuel cells,” in Inspec, Compendex, NTIS and Referex.
Data gleaned from the facets feature may be charted on a bar chart and/or downloaded.
In addition, the statistical data from any facet can be exported into an Excel spreadsheet for more sophisticated manipulation and graphing ( these exports are presented in .TAB format, which can be opened with Microsoft Excel). The export data can be generated for any Facet by selecting the small Microsoft Excel icon that appears next to each facet heading. The figure below shows the result of an export on the author data from a keyword search on “fuel cells,” in the four collections mentioned above.
As mentioned earlier, the data points that are included in the facets bar may change depending on what files are being searched. If a combination of files are searched, data points in the facets will stick to the very basic fields shared by all files (e.g. author, publication date, etc.). However, if searches are restricted to files with special indexing, the facets bar may include these special fields as options for refining the search. For example:
- When a search was restricted to Compendex, the facet bar included “Controlled vocabulary” and “(Compendex) Classification code” as refining options
- When a search was conducted in Referex, facet options included “Book Collection” (Referex is divided into a number of collections by subject,), and “Keyword.”
(This effect can also be observed when exclusively searching in other Engineering Village files as well – this list is not exclusive. )
Editor's Note:The Facet bar has three main functions that are highly beneficial to serious searchers. This feature:
1. Provides instant statistical summaries by listing or graphing the top data points and their frequency counts
2. Provides a fast way to refine data sets by excluding unwanted sub-sets
3. Provides enhanced iterative searching and drill-down investigations
While points 1 and 2 are somewhat obvious given the description above of the facet bar’s functionality, point 3 is somewhat more abstract. However, it will be immediately obvious to experienced searchers that facets can increase the speed of iterative searching tremendously.
For example, say that a user is interested in fuel cells, so he performs a keyword search in Engineering Village. The searcher can immediately see from the facets bar that the most common author affiliation for his results set is Argonne National Lab, in Illinois. By selecting this data point from the facets bar, the user restricts the results set to only those results from Argonne National lab. The facets bar re-sets to show only the data points for this restricted sub-set, and it is immediately apparent that the top three inventors at this institution are M. Krumpelt, G. H. Kucera, and R. D. Pierce.
Without the facets bar, this kind of statistical data about results sets is not readily apparent. If the searcher had approached this through a platform without filtering capabilities, (s)he may have needed to export and graph the data to see these top performers. Or perhaps the searcher would not have even considered this data, were it not immediately available. A huge benefit to the facets bar is that it can act as a source of inspiration for searchers to follow the type of “bunny trails” that often arise during searches, where new search ideas become apparent as the users progress through an initial results set.
Finally, although the graphs produced by the facets bar are not interactive, some users may find them useful for on-the-spot visualization. If professional –quality graphs of this data are needed, the data export function can be used to export the data into Microsoft Excel or another visualization program which can handle the data format.
