Conversational interfaces recently gained a lot of attention. One of the reasons for the current hype is the fact that chatbots (one particularly popular form of conversational interfaces) nowadays can be created without any programming knowledge, thanks to different toolkits and socalled Natural Language Understanding (NLU) services. While these NLU services are already widely used in both, industry and science, so far, they have not been analysed systematically. In this paper, we present a method to evaluate the classification performance of NLU services. Moreover, we present two new corpora, one consisting of annotated questions and one consisting of annotated questions with the corresponding answers. Based on these corpora, we conduct an evaluation of some of the most popular NLU services. Thereby we want to enable both, researchers and companies to make more educated decisions about which service they should use.
An approach to generating 'intelligent alarms' is presented that aggregates many information items, i.e. measured vital signs, recent medications, etc., into state variables that more directly reflect the patient's physiological state. Based on these state variables the described decision support system AES-2 also provides therapy recommendations. The assessment of the state variables and the generation of therapeutic advice follow a knowledge-based approach. Aspects of uncertainty, e.g. a gradual transition between 'normal' and 'below normal', are considered applying a fuzzy set approach. Special emphasis is laid on the ergonomic design of the user interface, which is based on color graphics and finger touch input on the screen. Certain simulation techniques considerably support the design process of AES-2 as is demonstrated with a typical example from cardioanesthesia.
%DVHG RQ DQ H[SHUW ¶V GLJLWDO WUDFH LQ D FRPSDQ\ semantic technologies in combination with enterprise social media applications enable the identification of experts in a required field of knowledge. We analyze how this identification of experts can be used to target the right crowd for corporate problem solving and analyze how accurate this expert identification algorithm can be in the case of sparse digital trace data. The case study presents real-world data of the so-FDOOHG 8UJHQW 5HTXHVWV IURP 6LHPHQV ¶ 7echnoWeb, a Siemens-internal crowd sourcing method. The three metrics spam reduction factor, gain factor and conversion rate are defined in order to measure the quality of the semantic message targeting in relation to a simple broadcasting of the Urgent Request to every TechnoWeb user. We apply these metrics to the real-world data of the Urgent Requests and analyze the reasons why some messages are better targeted than others.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.