Imagine you’re browsing through your social media or news feed, endlessly scrolling while looking for something to ignite your imagination. You’re on the lookout for that little hint of an emerging idea before it explodes into an entire industry. You are experiencing the silent frustration of a knowledge worker or researcher or simply someone who enjoys learning, as you are bombarded by too much information. But then you come across a new feed than those you’ve previously encountered. Instead of being a static flow of content and news; this new feed is more like a static map of all that’s happening, as it relates to research, ideas, opportunities and developments from around the world. This is what Wispaper offers you; not by merely compiling the “popular” things people are talking about, but by proactively mapping what’s currently happening, as a whole, and what’s about to occur next. Wispaper uses a simple, but powerful, premise; the best sources for innovative new ideas are NOT typically the loudest. Most often, they’re the soft / quiet (whisper) signals being created, through a collaborative process of sharing between disciplines, and waiting for their time to be amplified. In what way does wispaper evolve from just being a basic search engine / alertness system, into a proactive scouting mechanism for emerging trends in academic research? It is a combination of smart infrastructure, semantic comprehension, and a design philosophy that values discovery rather than just retrieving.
The Wispaper Engine: Beyond Keywords to Conceptual Landscapes
The cycle of traditional research tools can confine you into the trap of your own making when you enter in a search term, receive results with papers that contain that term, and in turn further entrench your “knowledge map.” The wispaper product breaks down that echo chamber and builds a real-time, evolving conceptual landscape of scholarly research. The wispaper software not only captures the traditional citation and keyword patterns; it looks deeper into the semantic aspects of research (e.g., methodology, central problems being addressed, subtle shifts in terminology to signify evolving concepts). When you use wispaper, you are not only setting up an alert for your interest in “quantum machine learning,” but you are beginning to tune into the entire ecosystem around this concept. In evaluating the data collected by wispaper, the system is also able to determine which sub-problems are gaining momentum within the traditional disciplines and where unexpected methodological crossovers are occurring. This information enables wispaper to identify trends that are in their infancy; in fact, some trends may be identified prior to the establishment of a distinct keyword or prior to any noticeable growth in the number of publications associated with those trending concepts. For example, an initial trend may arise from finding new ways of implementing common methodologies from another area of research that use data not recognized (metaphorically) when being tracked by search engines or keyword crawlers but are instead recognized when applying contextual insights to look for additional, more relevant information. Through this type of methodical mapping out of these latent or hidden linkages wispaper has successfully produced a continuously updated map/repository of new and additional trails identified via the above process that point towards areas yet to be explored.
The proactive establishment of services comes from a commitment to ongoing deep learning and the creation of models capable of detecting emerging patterns. Are certain mathematical models suddenly popping up in biology journals after being absent for decades from the field of engineering? Or, are researchers in neuroscience and computer science now citing a set of common foundational theory papers out of philosophy? These are the types of breadcrumbs that wispaper tracks. The way wispaper does this is not just a simple count of citations, but by analyzing both the pattern of citations and the contextual factors related to those citations. A more pronounced and diverse increase in citation activity within multiple unrelated disciplines can be interpreted as a stronger signal than a slow, incremental growth of citation activity within one discipline. Your personalized feed is made up of a collection of these signals that have been detected and contextualized against a list of papers endorsing that event and style of convergence. Your feed will not say, “Here are ten papers on X.” According to wispaper, “This is a growing conversation between disciplines A and B that focuses on the idea C, and you can find the most exciting contributions to this conversation over the last thirty days in 5 of these papers.” The result is that you are not just an observer but also a pioneer at the edge of something new.
Your Feed, Your Frontier: Personalization Without Isolation
A frequent error made by most recommendation systems is to generate an ideal, yet paralysing filter bubble. If you only ever look at papers on computational linguistics, then a naive recommendation engine would simply output lots more papers about computational linguistics. Wispaper avoids this pitfall by determining a difference between what is core to your interests and what is peripheral to your exploration. Your feed is created with a two-layer model of discovering new content. The first layer represents the more traditional area of academic focus-that is, a continuous stream of substantive output in your primary area of study, and filtered through the wispaper lens based upon true novelty as opposed to simple recentness. The second and potentially most important layer is the adjacency layer, where the advantages of wispaper’s proactive trend-spotting become evident. The adjacency layer considers the following question: Given this user’s particular area of interest, what kinds of discussions are happening now in an adjacent area of study that could be significantly impactful in 6 months with respect to user’s core area of research? The answer may not always be apparent, but the use of wispaper’s semantic mapping provides enough of a basis for the potential to be valid.
Imagine that you work as a materials scientist researching perovskite PVs (Photovoltaic). Wispaper will certainly track new developments in stability and efficiency, but it may also bring you new research, such as a preprint of a new AI-method for protein folding simulation in computational biology. What is the link? Both are complex, multi-dimensional optimisation problems that take place in unstable environments; thus, the AI-method presented in the biology article may be applied to understand the degradation of perovskites. You would not have actively sought out this information, but wispaper has identified the connection between the methods and brought it to you. So now your feed becomes your own personal frontier: the known world of your specialisation meets the unknown wilderness of cross-disciplinary possibilities. The platform provides you a trusted companion as you traverse the paths from your home territory to exciting, applicable areas of research. Therefore, you are not only up to date with your field, but also are prepared to explore new ideas that may change the way we see your field due to external influences.
The Human-in-the-Loop: Serendipity Engineered
Wispaper’s success in helping people identify trends lies largely in its human-in-the-loop design approach. Although Wispaper uses this principle in operation, it also takes care to design for incidental discovery. Serendipity is often thought of as chance; however, by developing a mechanism by which we can structure opportunities for serendipity, Wispaper creates the potential for unintentional discoveries. Wispaper provides two primary features that create structured opportunity for serendipitous discoveries: Concept Bridges and Methodology Crossroads. These features provide information not only by listing papers but also visually demonstrate how tools and ideas are moving across scholarly silos via the use of “Concept Bridges” and “Methodology Crossroads”. By displaying this information visually and contextually, Wispaper allows your brain-the most powerful tool available to you-to make connections and recognize patterns. Therefore, if you were to log into Wispaper to review a specific trend and find yourself exploring a concept map and discovering an unknown node, it could take you down a path you never expected; however, it may ultimately result in a major change in your perspective.
The freeform and less formal feel of the platform is where it shines. Rather than reading a textbook, you’re reading something akin to a lab journal where people are sharing experiments they conducted with each other; the summary will do more to explain why it matters than merely stating an abstract. Terms like, “If you’re going to investigate this method of doing this type of work, you may want to look into how it applies to other areas of research,” help foster connections and reduce the amount of work needed in order to enter into an unknown field at the same time. Wispaper is trying to create a similar experience as you had during a conference break with colleagues in a completely different field; while discussing what someone told you at the coffee stand might trigger an idea for you to try in your own work. Wispaper will provide you with an ongoing flow of these types of connections from multiple disciplines into your feed.
From Signal to Insight: The Wispaper Workflow
The ultimate test will be how much wispaper helps you to be successful with the workflows created by using those trends. Simply receiving signals about new trends before they become more mainstream will not benefit you unless you are able to incorporate those signals into your workflow for consuming, analyzing and synthesizing knowledge. Wispaper was built to be your beginning point (not your end point). Each trend pulled from your wispaper feed is a portal into gathering additional information. Within wispaper, integrated tools enable you to follow citation threads to see the authors who have collaborated on this work, as well as the bibliographic information for that work, in an accelerated manner. With all of this information being available to you, you can move quickly from identifying a trend to understanding who has developed that trend and what they are based on, with as little effort as possible. This is a way to manage the data overload created by academic publishing and create access to concise and contextualized data.
While there may be countless sources to get information from, they only provide very little of it in an insightful manner. Wispaper is not just a way to aggregate that information, but also a means by which we will assist in finding it through proactive means, anticipating trends through semiotic intelligence. Essentially, this represents a fundamental shift from searching for something we already know we need, when what we really should be searching for is something we have yet to discover may be truly valuable. Instead of being “just” a bulletin board that primarily displays what exists today, the feed will become increasingly like a radar screen identifying the very first signs of the next big thing, ready for someone with an inquisitive mind to pursue further. This transformation will ultimately allow the platform to create a connected experience as well as provide guidance to users on their journey to the far corners of what is known.
