Will This Idea Spread Beyond Academia? Understanding Knowledge Transfer of Scientific Concepts across Text Corpora
32321Authors
nan
Abstract
AbstractWhat kind of basic research ideas are more likely to get applied in practice? There is a long line of research investigating patterns of knowledge transfer, but it generally focuses on documents as the unit of analysis and follow their transfer into practice for a specific scientific domain. Here we study translational research at the level of scientific concepts for all scientific fields. We do this through text mining and predictive modeling using three corpora: 38.6 million paper abstracts, 4 million patent documents, and 0.28 million clinical trials. We extract scientific concepts (i.e., phrases) from corpora as instantiations of “research ideas”, create concept-level features as motivated by literature, and then follow the trajectories of over 450,000 new concepts (emerged from 1995-2014) to identify factors that lead only a small proportion of these ideas to be used in inventions and drug trials. Results from our analysis suggest several mechanisms that distinguish which scientific concept will be adopted in practice, and which will not. We also demonstrate that our derived features can be used to explain and predict knowledge transfer with high accuracy. Our work provides greater understanding of knowledge transfer for researchers, practitioners, and government agencies interested in encouraging translational research.
Keywords
Full Content
Findings of the Association for Computational Linguistics: EMNLP 2020, pages 154–163 November 16 - 20, 2020. c⃝2020 Association for Computational Linguistics 154 Converting the Point of View of Messages Spoken to Virtual Assistants Isabelle G. Lee Vera Zu Sai Srujana Buddi Dennis Liang Purva Kulkarni Jack G.M. FitzGerald Amazon Alexa { isabelll, xinzu, buddisa, liangxin, kulkpurv, jgmf }@amazon.com Abstract Virtual Assistants can be quite literal at times. If a user says tell Bob I love him, most virtual assistants will extract the message I love him and send it to the user’s contact named Bob, rather than properly converting the message to I love you. We designed a system that takes a voice message from one user, converts the point of view of the message, and then delivers the result to its target user. We developed a rule- based model, which integrates a linear text clas- sification model, part-of-speech tagging, and constituency parsing with rule-based transfor- mation methods. We also investigated Neural Machine Translation (NMT) approaches, in- cluding traditional recurrent networks, Copy- Net, and T5. We explored 5 metrics to gauge both naturalness and faithfulness automatically, and we chose to use BLEU plus METEOR for faithfulness, as well as relative perplexity using a separately trained language model (GPT) for naturalness. Transformer-Copynet and T5 per- formed similarly on faithfulness metrics, with T5 scoring 63.8 for BLEU and 83.0 for ME- TEOR. CopyNet was the most natural, with a relative perplexity of 1.59. CopyNet also has 37 times fewer parameters than T5. We have pub- licly released our dataset, which is composed of 46,565 crowd-sourced samples. Introduction Virtual Assistants (VAs), such as Amazon Alexa or Google Assistant, are cloud-based software systems that ingest spoken utterances from individual users, detect the users’ intents from the utterances, and perform the desired tasks (Kongthon et al., 2009; Tunstall-Pedoe, 2014; Memeti and Pllana, 2018). Common VA tasks include playing music, setting timers, answering ency- clopedic questions, and controlling smart home devices (L´opez et al., 2017). In addition, VAs are used for com- munication between two users. With the help of VAs, users can make and receive calls, as well as send a voice message or text message to their contacts. This paper focuses on voice messaging. Messaging can be implemented in such a way that the messages are snipped from the utterance in their spoken form (Mohit, 2014). For example, a user may say tell Bob that I’m running late. The Named Entity Recogni- tion (NER) model could extract I’m running late as the message content and pass that to the recipient directly. Such an approach works in many cases, but it does not perform well if the user says something like ask bob if he’s coming for dinner, for which the recipient would receive (if) he’s coming for dinner using a simple NER snipping approach. In this way, direct messages are dis- tinguished from indirect messages (Li, 1986). Indirect messages require further natural language processing (NLP) to convert the point of view (POV). Crucially, the …
Content truncated to 500 words
Labeled Text
AbstractWhat kind of basic research ideas are more likely to get applied in practice? There is a long line of research investigating patterns of knowledge transfer, but it generally focuses on documents as the unit of analysis and follow their transfer into practice for a specific scientific domain. Here we study translational research at the level of scientific concepts for all scientific fields. We do this through text mining and predictive modeling using three corpora: 38.6 million paper abstracts, 4 million patent documents, and 0.28 million clinical trials. We extract scientific concepts (i.e., phrases) from corpora as instantiations of “research ideas”, create concept-level features as motivated by literature, and then follow the trajectories of over 450,000 new concepts (emerged from 1995-2014) to identify factors that lead only a small proportion of these ideas to be used in inventions and drug trials. Results from our analysis suggest several mechanisms that distinguish which scientific concept will be adopted in practice, and which will not. We also demonstrate that our derived features can be used to explain and predict knowledge transfer with high accuracy. Our work provides greater understanding of knowledge transfer for researchers, practitioners, and government agencies interested in encouraging translational research.
Publication Details
- Paper ID:
- 32321
- Year:
- 2020
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan