Attended Day 1 (Monday May 5th) of 2003 PIUG Annual Meeting


Abstract
The 2003 Patent Information Users Group (PIUG) Annual Meeting was held in Chicago with the theme of Patent Analysis. Day One of the meeting featured sessions on Patent Analysis for R&D Strategic Technical Planning and Patent Analysis for Legal Groups. Talks were presented by participants from Vertex Pharmaceuticals, WHU, Rohm & Haas, 3M, Bristol-Myers Squibb, Ciba Vision, The IPI Institute, Quisitor, Pfizer and the European Patent Office.

Last year, during the 2002 PIUG Annual Meeting, Lucy Akers, PIUG Chair, told me that the organization was considering selecting patent analysis as the theme of the 2003 meeting. Being a topic that is near and dear to my heart I was of course very excited about this and immediately volunteered to participate in the meeting. As it turned out, over the course of the subsequent year I was completely unprepared for the amount of work that was going to be required to help pull something like this together. Even though it has been a lot of work on the part of a great many people, I am happy to report that the 2003 PIUG Annual Meeting has started very successfully and a great collection of talks were presented during the first day.

Some thoughts on the presentations given during day one are discussed below:

Anthony Trippe, Vertex Pharmaceuticals: Patent Analysis the Once and Future Discipline

I was extremely honored to be asked to present the keynote talk to this year's meeting. I will cover my talk in more detail in the New Presentations and Papers category at a later date so I will simply comment again here that I was very grateful to be asked to give this talk.

Prof. Dr. Holger Ernst, WHU Chair of Innovation and Technology Management: Using Patent Information for Strategic Technology Management

I discovered Dr. Ernst's (he prefers Holger but I just seem to always end up calling him Dr. Ernst) papers in one of the SDIs I have set up on patent analysis. Reading his papers I was very impressed by his understanding of patent information and his grasp of some of the nuances and challenges inherent in using patent information in statistical exercises. I also appreciated the mathematical and scientific approach he brings to this field and the fact that he seems to understand the underlying nuances of patent data better than some of his colleagues.

Dr. Ernst's talk started with an explanation of several metrics his group has developed or implemented for studying patent information. Particularly interesting were his discussions of patent share (which is directly analogous to market share) and patent strength which is a measure of the quality associated with a particular document. He was quick to point out that use of these metrics on the micro-level scale of individual patents is not nearly as useful (or accurate) as using them for large, macro-level projects.

Holger went on to discuss a number of examples of how the metrics could be applied to different situations in the strategic technical planning process. Particularly interesting was a new framework he demonstrated that married patent share (properly weighted using the patent strength of the participants) with patent growth and market share. The addition of the market date has apparently helped a number of European companies make decisions on product develop and marketing strategies. Finally (the talk ran a little into the questions and answers period), Holger showed us an example using inventor analysis demonstrating the importance of identifying, and perhaps more importantly, rewarding inventors who make an impact on a business.

Kerryn Brandt, Rohm and Haas: Roles for the Search Professional as a Partner in Patent Analysis

As I mentioned in my opening comments when introducing Kerryn, I have known him for years and have been asking him for some time to speak at one of my symposiums on the role he has played bringing text mining and analysis tools to his company. Finally, Kerryn agreed to speak at the 2003 PIUG Annual meeting and I can say that the result was worth the wait. Kerryn started by discussing the unique skills and experience that patent information professionals have with regards to requirements organizations need when selecting and using patent analysis tools. I especially liked his ideas with respect to a "food pyramid" for patent analysis. This pyramid has the "subject" has a base and is built up with the "data" and finally the "tool" at the top. Kerryn also cautioned about the "inverted pyramid" where companies start with a tool and try to apply the data and subject after the fact. Since the tool is the tip of the pyramid this arrangement leads to an unstable configuration. This is more or less analogous to what I have been saying with the linear law of patent analysis where the choice of tool should be the last decision made during an analysis project. I really like Kerryn's use of the pyramid as a visual for this assertion.

Kerryn went on to discuss three different, brief instances where the use of patent analysis tools has proven beneficial to his organization. He concluded by reiterating the important role information professionals play in the patent analysis tools selection process. Information professionals are ideally suited to provide an unbiased view into the different tools which can not be supplied by the vendors.

Richard Kurt, 3M: Data Visualization - Experiences and Lessons Learned at 3M

Richard provided an overview of some of the tools in use at 3M. They included, ThemeScape, VantagePoint, Infosleuth/Techtracker (which apparently offered some unique features but unfortunately is no longer available since the company went out of business shortly after 3M made their purchase), Hyperbolic tree visuals from InXight and online ranking tools such as the analyze command on STN International. Richard outlined several advantages 3M believes it gains from the use of these tools. They include, the ability to rapidly analyze large collections of text (since most users have a very low tolerance for working with large data sets), the top-down view and organization that the tools provide, the ability to formulate hypothesis and rapidly provoke discussion on the subjects and the strategic insights that the tools provide. Richard also cautioned that the visualizations are not a replacement for human analysis and that the field is still in it's infancy and further work will need to be done in the future. He concluded by challenging the assembled audience to step up and be a hero within their organization by shepherding these tools in their companies. Richard felt that his involvement in this area had certainly increased his standing within 3M.

Bruce Mason, Ciba Vision: Is There a Role for Patent Analysis in Competitive Intelligence?

Bruce made a number of good points with regards to the use of patent analysis in traditional competitive intelligence exercises. Since patent data is not normally available until, at least 18 months after the patents have been filed for some industries this lack time is an eternity. It is also true that some industries do not rely on patent protection as much as they do on other forms of protection such as trademarks and trade secrets. Bruce also made a point of demonstrating the use of Porter's Five Forces model for looking at the competitive environment surrounding companies. Patent data can be used in helping to represent some of these forces but are not useful for others. For the most part Bruce down played the importance of patent analysis with regards to other methods used to analyze the the competitive landscape and challenges for some companies and industries. He also used Procter and Gamble and Lucent as examples of companies where a Five Forces model could possibly help explain their performance.

Heahyun Yoo, Bristol-Myers Squibb: Enhancing the Quality of Biotech Patent Information with Sequence Analysis

Heahyun focused on the importance of conducting biosequence searches to identify important biotech patents that might not otherwise be located using more traditional searching methods. She provided an overview of the growth in biosequences in both the journal and patent literature and discussed several examples where vital patent documents were discovered via a sequence analysis.

Lucy Akers and Ford Khorsandian, The IPI Institute: A High Level Overview of Patent Search Types and Approaches

Lucy and Ford did a nice job of breaking down the different types of searches traditionally conducted by patent information professionals and discussing the relative importance and level of detail needed when conducting them. The four types of searches discussed were State-of-the-Art, Patentability, Validity and Freedom-to-Practice/Operate. Using an umbrella/weather metaphor, Lucy and Ford outlined the requirements for each search and demonstrated the business applications of each.

Julia Deboys, Quisitor Ltd. UK: Decision Pathways in Patent Searching and Analysis

There has been some agreement that there are at least two types of patent analysis, macro-level which deals with thousands of records and is done using mining techniques and visualization and micro-level techniques which focus on smaller document sets and require detailed examination. Julia's talk outlined an interesting method for conducting micro-level analysis using decision pathways as a means for prioritizing documents to be read by the client or analyst. Julia started by discussing the traditional use of decision pathways for making business decisions and demonstrated how the method produced valuable results. She concluded by applying the method to a patent search and showed how the process could be used to winnow an answer set down to those answers which were most likely to be of the highest interest to the client.

James Grant and Karl Neidert, Pfizer: Dominating Claims and Patent Analysis

James provided a practical example using Virocillin and capsule formulation to demonstrate how backward and subsequent forward United States Patent citation analysis helped produce a more comprehensive Freedom-to-Operate search than searching with keywords or chemical entities alone. James began with a more traditional formulation search using resources available in Chemical Abstracts and other chemical based files. His client, Karl Neidert, an attorney, was able to provide a more thorough Freedom-to-Operate opinion when citation analysis was also used since citation analysis is not hindered by the use of a specific keyword or compound strategy and will provide answers that are similar to what is specifically being requested. In this way answers that might appear obvious but are not exactly what is being searched for can be reviewed.

Annemie Nuyts , European Patent Office: EPOQUE and the Suite of Examiner's Applications at the European Patent Office

Annemie provided an inside look at how examiners at the European Patent Office conduct searches using the EPOQUE system and what tools the examiners had at their disposal. At the EPO a combination of citation, classification and keyword searching is done to produce the most comprehensive results. The examiners also have at their disposal a system for ranking the results and drilling into the individual patent documents under review.

Posted: Mon - May 5, 2003 at 02:41 PM   Patinformatics   Tony's Reflections   Email Comments

© Anthony Trippe