Accelerating Innovation Through Analogy Mining

Abstract

e availability of large idea repositories (e.g., the U.S. patent database) could signi cantly accelerate innovation and discovery by providing people with inspiration from solutions to analogous problems. However, nding useful analogies in these large, messy, realworld repositories remains a persistent challenge for either human or automated methods. Previous approaches include costly handcreated databases that have high relational structure (e.g., predicate calculus representations) but are very sparse. Simpler machinelearning/information-retrieval similarity metrics can scale to large, natural-language datasets, but struggle to account for structural similarity, which is central to analogy. In this paper we explore the viability and value of learning simpler structural representations, speci cally, "problem schemas", which specify the purpose of a product and the mechanisms by which it achieves that purpose. Our approach combines crowdsourcing and recurrent neural networks to extract purpose and mechanism vector representations from product descriptions. We demonstrate that these learned vectors allow us to nd analogies with higher precision and recall than traditional information-retrieval methods. In an ideation experiment, analogies retrieved by our models signi cantly increased people's likelihood of generating creative ideas compared to analogies retrieved by traditional methods. Our results suggest a promising approach to enabling computational analogy at scale is to learn and leverage weaker structural representations.

Multi-Vector Models with Textual Guidance for Fine-Grained Scientific Document Similarity

“…An interesting direction to explore is the interactive use of our methods, with a system allowing users to highlight specific aspects of papers and retrieve contextuallyrelevant matches. Another interesting application is using our model for finding analogies -structural matches between texts describing ideas, such as scientific papers -to boost discovery (Hope et al, 2017(Hope et al, , 2021bChan et al, 2018).…”

“…Our models, unlike previous work (Neves et al, 2019;Jain et al, 2018;Hope et al, 2017), make no assumption on specific aspect semantics in deriving a model architecture, and focus on aspects in the form of general subsets of document sentences.…”

“…The ability to identify similarity across documents in large scientific corpora is fundamental for many applications, including recommendation (Bhagavatula et al, 2018), exploratory or analogical search (Hope et al, 2017(Hope et al, , 2021bLissandrini et al, 2019), paper-reviewer matching (Mimno and McCallum, 2007;Berger et al, 2020) and many more uses.…”

Bursting Scientific Filter Bubbles: Boosting Innovation via Novel Author Discovery

“…This cycle of input, updating, engagement, and response can lead to an amplification of biases around searchers' prior awareness and knowledge [16]. Such biases include selective exposure [9], homophily [21], and the aversion to information from novel domains that require more cognitive effort to consider [13,17]. By reinforcing these tendencies, algorithmic systems that filter and rank information run the risk of engendering so-called filter bubbles [28] that fail to show users novel content outside their narrower field of interest.…”

“…Scientific filter bubbles can be costly to individual researchers and for the evolution of science as a whole. They may lead scientists to concentrate on narrower niches [18], reinforcing citation inequality and bias [27] and limiting cross-fertilization among different areas that could catalyze innovation [13]. Addressing filter bubbles in general, in domains such as social media and e-commerce recommendations, is a hard and unsolved problem [6,10,48].…”