Understanding and improving information search – a cognitive approach
The term ‘cognitive search’ is commonly used to label how a search application can read the minds of searchers as to their intent in undertaking a search. This book, Understanding and Improving Information Search – a Cognitive Approach takes a much wider perspective, considering the entire process of search, especially exploratory search. I am normally disappointed by the consistency and currency of multi-author books but this title has been assembled with great skill and covers research published in 2019. This is immediately obvious with the introduction by the co-editors Wai Tat Fu and Herre van Oostendorp which leads the reader through the book with great care.
The first part of this book focuses on the computational cognitive modeling framework that integrates information retrieval metrics into cognitive simulations of user behavior in the broader information search process. The opening chapter is a tour de force from Peter Pirolli building on the work that he and Stuart Card undertook in the late 1990s on information foraging, followed by a chapter from Wai Tat Fu on How Cognitive Computational Models Can Improve Information Search.
Having laid the foundational framework, in the second section there are five chapters on the methods and tools that support the cognitive approach. I must highlight Chapter 7 entitled Designing Multistage Search Systems to Support the Information Seeking Process which highlights the importance of designing search user interface features for different information seeking stages (building on the work of Kuhlthau and Wilson amongst others) and should convince you that everything you thought you knew about user interface design for search is probably wrong. In Chapter 8 Kazutoshi Umemoto, Takehiro Yamamoto and Katsumi Tanaka go into further detail on UI design issues. Building on the work of Marti Hearst the authors present some very elegant UI solutions for relevance judgment, information credibility, exploratory search and accommodating a range of search skills. If you are of the opinion that the ‘modern’ UI on Microsoft 365 Search cannot be improved then you may want to give this chapter a miss.
The final section of this 286page book considers a few topics in more depth, of which Chapter 13 on Conversational Interfaces for Information Search is especially valuable given the current level of interest in chat bots.
Although this book is grounded in academic research it is in general written for a more general audience. Enterprise search developers, especially those involved in user interface development, will find much to challenge and inspire them, and this would also be the case for search managers in organisations with a decent level of enterprise search maturity. Inevitably the chapters focus on web search (when will the academic community realise the vast differences between web and enterprise search!) but the general principles expounded in this book are quite widely applicable to enterprise search implementations by anyone with a commitment to understanding how to deliver high levels of search satisfaction. It should also appeal to information scientists, especially those with a strong interest in the psychology of information behaviour and information search.
This book is relatively expensive at £87.50 but for me it has been a Rosetta Stone journey on reading through it as many of the observations I have made over the last decade now start to make sense when considered within a broad cognitive experience agenda. Those who claim that ‘cognitive search’ is a solved technology problem will read this and appreciate that a) there is so much research they should be aware of and b) cognitive search is about understanding user behaviours at a very detailed level.