Classically, training relation extractors relies on high-quality, manually annotated training data, which can be expensive to obtain. To mitigate this cost, NLU researchers have considered two newly available sources of less expensive (but potentially lower quality) labeled data from distant supervision and crowd sourcing. There is, however, no study comparing the relative impact of these two s...

Building accurate knowledge graphs is essential for question answering system. We suggest a crowd-to-machine relation extraction system to eventually fill a knowledge graph. To train a relation extraction model, training data first have to be prepared either manually or automatically. A model trained by manually labeled data could show a better performance, however, it is not scalable because a...

We present a newly crowd-sourced data set of natural language references to objects anchored in complex urban scenes (In short: The REAL Corpus – Referring Expressions Anchored Language). The REAL corpus contains a collection of images of real-world urban scenes together with verbal descriptions of target objects generated by humans, paired with data on how successful other people were able to ...

In view of some problems existing in traditional software crowdsourcing tests and Internet Things device tests, we developed a CrowdIoT test system for the devices based on block chain. uses core technology blockchain to focus its testing physical Things. two different methods needs: online offline testing. Online remote is our key function research object. By opening interface testers, testers...

In order to support advanced collaborations among smart products, services, users and service providers in a product ecosystem (S-PSS), this paper proposed service-oriented hybrid digital twin (DT) thread platform-based approach with embedded crowd-/service-sourcing mechanism for enabling manufacturing services. This is well supported by the interaction intelligence of digitally connected users...