Research Outputs and Analytics

InCitesTM Use Cases

InCitesTM Benchmarking & Analytics is of value to researchers and administrators. It is useful for finding expert collaborators or prospective faculty recruits, analyzing institutional strengths, identifying existing partnerships with industrial or international scope, and many more. Here, we have collected a few of the most common questions asked and provide a step-by-step walk-through of how InCites can provide answers.

  1. Since you are looking for UGA faculty, choose the People tile on the InCites home page
  2. Hide the visualization while adjusting filters
  3. First choose a time period for the analysis; a good starting point is to limit to the last 10 years (2007 – 2016, currently); under Tile Settings | Time Period, drag the left marker until it reads Min: 2007
  4. You want publications with author affiliations of UGA, so under Filters | By Attributes | Affiliated Organization, start typing University of Georgia, then select when it comes up in the list
  5. Then you want just those UGA authors who have collaborations in a specific region; under By Research Network | Collaborations with Locations, start typing Sao Paolo, then select when it comes up in the list (note that different geographic regions may be present with more or less granularity; if your desired city/region is not in the list, perhaps try to identify a group of organizations instead)
  6. To apply these filters, click Update Results (bottom of left margin); the table will adjust. When this was run in early 2017, 153 results were generated (top of left margin)
  7. Sort the table by using column headings; in this case it might make most sense to sort on Web of Science Documents (this is a publication count); click on that column heading. If you want to drill down to the publication lists, just click on the number in this column for any person
  8. Now Show the visualization and adjust to illustrate the data you want to highlight. As an example, pick Collaborations style (leftmost dropdown), display Category Normalized Citation Impact (next dropdown), and choose 15 in the third window (meaning show me the top 15 records in the table). Note that the attribute you display and the attribute by which the table ( is sorted are independent of each other

Note that this list has four name variants of Dorothy M. Fragaszy that appear as different authors: Fragaszy, Dorothy; Fragaszy, D.; Fragaszy, Dorothy M.; Fragaszy, D. M. These results obviously should be concatenated into a single record in the table; this is not explicitly possible, but you can come close:

  • Check the box to the left of all four name variants in the table. Checking any boxes brings up a banner at the bottom of the window, in which you should click the Pin To Top button. This moves all the name variant results to the top of the table.
  • Then click Benchmarks (above the table header) and choose Baseline for Pinned Items from the menu. This creates the baseline for all four name variants at the top of the table, giving you all of Dorothy Fragaszy’s results in one row (but not affecting the visualization).
SanPaolo Analytics screen shot.
Click to enlarge.
  1. Since you are looking for UGA faculty, choose the People tile on the InCites home page
  2. Hide the visualization while adjusting filters
  3. First choose a time period for the analysis; a good starting point is to limit to the last 10 years (2007 – 2016, currently); under Tile Settings | Time Period, drag the left marker until it reads Min: 2007
  4. You want publications with author affiliations of UGA, so under Filters | By Attributes | Affiliated Organization, start typing University of Georgia, then select when it comes up in the list
  5. Then you want just UGA authors that have published in a given research area (AI). Click on By Research Output | Research Area. Web of Science publications have been categorized using a number of area taxonomies (schema). First, choose which Schema to use; Web of Science is a good granular one unless you already know another one that properly categorizes the area you are looking for
  6. Clicking in the Research Area textbox brings up the entire set of research areas in that schema. In this case, probably the best choice is COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
  7. To apply these filters, click Update Results (bottom of left margin); the table will adjust. When this was run in early 2017, 129 results were generated (top of left margin)
  8. Sort the table by using column headings; in this case it might make most sense to sort on Web of Science Documents (this is a publication count); click on that column heading. If you want to drill down to the publication lists, just click on the number in this column for any person
  9. Now Show the visualization and adjust to illustrate the data you want to highlight. As an example, pick Bar Graph style (leftmost dropdown), display Category Normalized Citation Impact (next dropdown), and choose 10 in the third window (meaning show me the top 10 records in the table). Note that the attribute you display and the attribute by which the table (and visualization) is sorted are independent of each other
Table of results
Click to enlarge.
  1. It may be useful to know whether these top researchers in AI collaborate with one another (i.e., coauthor these publications). In the leftmost dropdown of the visualization, choose Collaboration Network; in this case, 8 of the top 10 researchers form a fairly extensive coauthorship network (roll over donut segments), but Prashant Doshi and Hamid Arabnia are outside of that network
Al Network screen shot
Click to enlarge.
  1. Choose the Organizations tile on the InCites home page
  2. Hide the visualization while adjusting filters
  3. First choose a time period for the analysis; a good starting point is to limit to the last 10 years (2007 – 2016, currently); under Tile Settings | Time Period, drag the left marker until it reads Min: 2007
  4. In the left margin, under By Research Output | Research Area, choose the Web of Science Schema, then choose Women’s Studies in the list of research areas
  5. Click Update Results; in early 2017, this generated a list of 2,276 organizations globally
  6. Note that many of the organizations in the list are university systems, which appear in addition to their university components. Under By Atttributes | Organization Type, select Academic System from the list, then click the = sign, turning it into a ≠ sign. Click Update Results, which then will show organizations that are not Academic Systems
  7. Alternatively, you could choose only to look at Academic organizations by selecting Academic as Organization Type and clicking ≠ so it turns back to =, then clicking Update Results
  8. You may also wish to limit your search to US organizations; under By Attributes | Location, choose USA from the list. Clicking Update Results gives 548 results in early 2017
  9. Choose how to sort the results table, perhaps by Web of Science Documents. This ranks the organizations based on quantity of output. Alternatively choosing Category Normalized Citation Impact provides rankings based on a measure of quality. However, some organizations with only a few WoS Documents skew this ranking, so set a Threshold so that these organizations are not considered. In the Filters section on the left, click the Threshold box and under Thresholds | Web of Science Documents, click the 0 after Min to pick a new threshold (say, 50). Click Update Results; now the 106 remaining organizations are ranked based on a quality metric
  10. To investigate the authors at a given institution, click the institution name in the table, pick Affiliated People in the first dropdown, then click Preview. This previews a table of authors, which can also be sorted. Click the number under Web of Science Documents for any author to see those documents
  11. Show the visualization if you wish; for example, choose Treemap and Web of Science Documents
Click to enlarge.
  1. Choose the Organizations tile on the InCites home page
  2. Hide the visualization while adjusting filters
  3. First choose a time period for the analysis; a good starting point is to limit to the last 10 years (2007 – 2016, currently); under Tile Settings | Time Period, drag the left marker until it reads Min: 2007
  4. To limit your search to only private organizations, under By Attributes | Organization Type, choose Corporate in the list; click Update Results. In early 2017, this generated 427 results
  5. Now you want the UGA co-authors for these publications; under By Research Network | Collaborations by Organizations, start typing University of Georgia and select when it appears; click Update Results. In early 2017, this reduced the list to 80 corporate organizations
  6. Sort the table by Web of Science Documents (if necessary) to see the highest quantity of work
  7. To investigate the research areas for a given company, click the company name in the table, choose Research Areas in the first dropdown and click Preview. The resulting table can be sorted and you can drill down on # of Web of Science Documents, then click a publication title to see Web of Science detail, including authors and affiliations

Show the visualization, then pick Pie Chart from the first dropdown and Category Normalized Citation Impact from the next. Show the top 12 companies. This gives you a feeling about the average quality (recognition) of the work for these collaborations

graph sowing results
Click to enlarge.
  1. Since you are looking for faculty, choose the People tile on the InCites home page
  2. Hide the visualization while adjusting filters
  3. First choose a time period for the analysis; a good starting point is to limit to the last 10 years (2007 – 2016, currently); under Tile Settings | Time Period, drag the left marker until it reads Min: 2007
  4. In this search, you probably want to be selective (only the best) so choose filters that are restrictive. First, under By Research Output | Research Areas, choose Web of Science Schema and CELL & TISSUE ENGINEERING as Research Area
  5. Focus on journal publications, so under By Research Output | Document Type, pick Article from the list
  6. Start by looking in this country, so under By Attributes | Location, choose USA from the list
  7. Click Update Results; in early 2017, this resulted still in 27,454 people, too many to consider
  8. You are looking for high-quality faculty with above-average Category Normalized Citation Impact values, so sort the table on that column; many faculty with a single publication in this area are near the top, but these will not be considered, so you need to set a threshold to only consider people with at least 10 (for example) Web of Science Documents. Click Threshold and reset the Min from 0 to 10
  9. Click Update Results; in early 2017, this resulted in 174 people, a much more manageable number
  10. Show the visualization and choose Bar Graph, Web of Science Documents, and top 10 (based on table sort: Category Normalized Citation Impact)
  11. This much is shown in the image; from here, you may want to limit to a region of the US (Location has states) or this could be a list of names for a recruiting committee to look through.

(Note that often authors are repeated, usually because they have multiple affiliations, either at the same time, or sequentially. See the first use case for how to group these records, pin them, and calculate the average baseline.)

Click to enlarge.
  1. Often, you need to filter the publication dataset to research areas that are more narrowly focused than the ones available in any of the InCites research area schema  This can be done by performing a keyword-based search in the Web of Science core collection to create a custom dataset focused on the research area of interest; this dataset can then be exported to InCites for further analysis
  2. The first task is to come up with a good set of keywords to be used in the search; in this case, consider terms like “mosquito”, “infectious”, “disease”. Then decide the appropriate Boolean combinations: “mosquito” .AND. “infectious” .AND. “disease”, meaning that any relevant publication must contain all three concepts
  3. It is also likely that you want to focus on recent work so choosing to limit to publications over the last 10 years (2007 – 2016) makes sense
  4. At UGA, the Web of Science (Web of Knowledge) Core Collection is one of the databases you can search in GALILEO. This is how this search would be constructed
screen shot of search page
Click to enlarge.
  1. When this search was run in early 2017, it generated 965 results; this seems like a reasonable number to start with for further analysis, but you may want to open up your search terms to obtain more results. InCites import is limited to no more than 50,000 publications
Screen capture of search page.
Click to enlarge.
  1. Near the top of the search results, choose Save to InCites from the dropdown menu, sign in to InCites (if you are not already), then choose a filename for your dataset and click Save
screen shot of how to save page
Click to enlarge.
  1. WoS will create your dataset and send you an email when it is available. When you receive that email, go to the InCites home page and click on My Folders (top right). You should see your new dataset; click on it and an Analytics page will appear
  2. You can choose Entity Type to analyze; if you choose People, you will see 5,221 results (these are all the authors of the 918 publications in your dataset that were imported; 47 of the 965 results were too recent to be in InCites yet)
  3. You only want to look at UGA-affiliated authors, so under By Attributes | Affiliated Organization, choose University of Georgia from the list; click Update Results
  4. In early 2017, this resulted in 12 UGA authors, which is a good start at identifying those who might be interested in this funding opportunity
Click to enlarge.