Our abstract proposal, Single Pass Numerical Matching A linguistic solution to numerical searching.” has been accepted by PIUG, so we will be off to Atlanta, Georgia in May. You can find the published abstract together with other the other speakers named so far by clicking the PIUG link above. You can see the full proposal by clicking on the title above.
II-SDV Conference 2017 Nice
We attended the 2017 II-SDV Conference in Nice, 24 - 25 April 2017. We are going to catch up with the latest developments and also taking with us our latest development, SpanMatch, which is what we will be talking about in Atlanta next month. This is a numerical search program specifically aimed at finding and calculating ranges against a given specification.
We developed it in association with David Goodchild, of David Goodchild Limited, whom we met in Madrid. We extended the initial English language version to cover Chinese, Japanese and Korean when we saw the state of some of the machine translations into English. You can find illustrations of it in use on the Patent Program tab.
We also discussed possible contributions we can make to the improvement of machine translation from our background in multilingual concordancing.
This was our first visit to an EPOPIC. We had done some consultancy work in the patent area and thought that our methodologies and presentation style would be of interest to the patent search community, so decided to go and ask them. In general the answer appears to be "Yes".
We attended Discussion Group 7, Freedom-to-operate, and were pleased to note at the outset that three items on the wish list of the questionnaire were met by our program, ability to highlight search phrases instead of only words, ability to save users' highlighting of text in descriptions and claims, and mumti-screen viewing to compare documents or parts of a document.
The discussion group was run in such a way that each table of 7 tackled 2 of the 5 assignments set by the chair, Susanne Hantos. As the designated note taker I had the privilege of hearing 12 professionals giving their take on the problems of handling a) geographical scope and b) inventions versus designs with great clarity and organisation in a time frame of 20 minutes for each team of 6.
From the conference as a whole it became clear that FTO was a very hard search task but equally a vitally important one for companies to ensure that their R&D efforts were geared towards products that had a chance of making it to market. It was also clear that, although there had been huge advances in machine translation, this was still an area where their was a lack of confidence in the accuracy of the current tools, particularly over extended passages.
As CFL works in both the document comparison area and multilingual concordancing we think we should be able to make further contributions to the power and accuracy of FTO search tools.