Image [cc] MysteryStevenson1

I ran across a summarizing tool that I think all of you would enjoy testing out. I found it through the Tools 4 Noobs website which focuses on PHP resources, but this can be used specifically from their “Summarize” tab on the website (no coding required… but, more on that below.)

I’ve run through a number of different web pages and copied PDF texts from Court filings, and found this summarizer to be pretty good.

Here’s the basic principles of what it does:

  1. Extract sentences from the text given.
  2. Identify the keywords in the text and count each word’s relevance.
  3. Identify the sentences with most relevant keywords and output them based on the options selected.

It also gives you some options to play around with to improve your results.

I took this PDF of a Plaintiff’s Original Complaint (7 pages), and got this as the summary:

Summary:

PRAYER For the reasons set forth above, Plaintiff respectfully prays that Defendants be cited to appear and answer herein, and for the following relief: a. that the Court enter an order allowing this action to proceed as a collective action under the FLSA and directing notice to the Workers; b. judgment awarding Plaintiff and the Workers all unpaid wages and unpaid overtime pay, liquidated damages, and attorneys’ fees and costs under the FLSA; c. an award of pre-judgment and post-judgment interest on all amounts awarded at the highest rate allowable by law; and d. all such other and further relief to which Plaintiff and similarly-situated employees may show themselves to be justly entitled.
Defendants failed to pay Plaintiff at the minimum wage and failed to pay Plaintiff at one and one-half time his regular rate of pay for hours worked in excess of forty (40) hours per each seven (7) day workweek.
Defendants’ practices of not paying Plaintiff and similarly-situated employees at the minimum wage and failing to pay at one and one-half times the regular rate of pay for all hours worked in excess of forty (40) hours per each seven (7) day workweek are direct violations of the FLSA.
Plaintiff and similarly-situated employees were entitled to overtime pay at one and one-half times their regular rates of pay for all hours worked in excess of forty (40) in a workweek.

For the IT folks, apparently you can purchase the coding for this ($60.00 for developer license or $15 for personal) and integrate it into the enterprise.

Looks pretty cool to me. Any ideas on what could be done with a product like this?? I was thinking how it could improve newsletters that Library staff create, or how about putting a summary like this in the DMS that allow us to see a summary of documents at a glance, rather than having to scroll through the preview (or worse, actually having to open the documents.)