FEATURE

Researchers are ready to develop search applications

Judson Dunham of Elsevier reveals the results of a recent survey into how researchers want to search for information and data

The continually evolving information landscape has presented an embarrassment of riches over the last few decades – and the scientific community has clearly benefited. However, with the amount of digitised scientific information increasing substantially and a rising trend in openly available raw data, scientists are spending too much valuable time digging through content.

This growing wealth of data has led to many discussions about information overload. However, the main problem isn’t too much information – researchers welcome the value of added content – but discoverability of the right information. Researchers want to consume more information but don’t want to sift through irrelevant data. Since the needs of researchers are as varied as the disciplines they represent, this is not a problem with a one-size-fits all solution.

Using the first-hand knowledge of those who can best identify the greatest search and discovery challenges – the researchers themselves – and empowering them to collaborate on solutions, may be the best way to offer personalised search and discovery. To foster researcher involvement, Elsevier recently launched SciVerse, a search and discovery platform that will encourage the creation and sharing of customised applications that meet specific researcher needs (see new products section).

During the development of this new product, Elsevier conducted qualitative and quantitative analysis into how researchers want to search. This indicated that researchers are interested in maximising the opportunities allowed by platform openness and interoperability. They are also seeking ways to leverage the broader web trends of personalisation and collaboration in the search and discovery process.

Quantitative studies

Elsevier followed this with a quantitative study of attitudes towards search technology trends. The company did a survey of more than 1,000 researchers across a wide range of disciplines, from around the globe, covering topics such as the potential impact of open data and the release of APIs (application programming interfaces). The findings overwhelmingly revealed that researchers are ready for the next phase in search technology and are more than willing to participate in bringing it to bear.

With respect to open data, the scientific community clearly recognises its potential to help accelerate the research process. Nearly all respondents to the Future of Search and Discovery survey agreed that open data is important to the future of search and discovery, with 71 per cent suggesting it is ‘very important’.

Despite this substantial demand, very little structure and organisation currently exists to facilitate easy and meaningful use of this data. In its current state, much of this content holds limited value to the scientific community. Reciprocal linking between published articles and corresponding data, as well as standard reporting formats, would enable greater interoperability and ease of use. However, to adequately access and truly gain added insight from this content, APIs will need to be made available so that custom applications can be developed for taking full advantage of this data.

The survey suggested that researchers are aware of this need. Nearly eight in 10 (78 per cent) of the researchers surveyed agreed that the availability of APIs will foster experimentation and the development of innovative search and discovery applications. By using available APIs to build applications that offer the most tailored solutions, researchers can improve their workflow and accelerate science. Ultimately, it is the availability of APIs that will provide the opportunity to meet the demand for delivery of personalised content and collaboration that the qualitative research showed the scientific community is seeking.

Interestingly, when asked which type of applications would be most useful to the scientific community, there was no clear winner. This may reflect the diverse needs of researchers, as well as the interest and potential to develop a myriad of applications. In fact, all of the application options in the survey had similar response rates as follows: applications that facilitate more customised search (18 per cent); those that extract data to elicit more meaningful insight (17 per cent); apps that show content which trusted peers find valuable (16 per cent); those that provide personalised content delivery based on my interests and background (16 per cent); and apps offering analytical tools that are able to target trends, look at historical research output and text/data mine to create semantic relationships across scientific content (16 per cent).

While the exact detail of which applications need to be developed first remains fuzzy, the endgame appears to be coming into focus. Nearly half (47 per cent) of respondents selected ‘collaborative knowledge networks (online groups of trusted peers)’ when asked which search technology trend will have the greatest impact on search and discovery over the next several years. Additionally, more than 80 per cent of respondents agreed that in the next several years researchers will use knowledge networks as a reliable source for filtering and viewing information.

The survey also uncovered a current enthusiasm for making this vision a reality. Almost seven in 10 researchers (68 per cent) expressed personal interest in developing search and discovery-related applications using scientific content for their institution. In addition, 61 per cent of those interested in developing such applications said their strongest motivation for doing so would be the ‘opportunity to help speed up research among the scientific community as a whole’. A more modest 32 per cent were most motivated by the desire to ‘speed up their own research.’ Even among those who did not express application development interest, two-thirds (66 per cent) said they would be more open to creating applications if there were an opportunity to collaborate with others who would handle the technical aspects of development.

As scientific data continues to grow at a rapid pace, and science itself becomes more and more multidisciplinary and collaborative in nature, one size-fits-all search and discovery is no longer enough. The landscape is changing and the scientific community is more than ready to participate actively in bringing in a new era in discoverability that will help address the specific pain points of researchers across the globe.

Judson Dunham is a senior product manager, Science and Technology at Elsevier

Feature

Cartoon abstracts are a very effective way to bring research to life, writes Daniel Pullin

Analysis and opinion