@misc{
 title = {Benchmarking natural language understanding services for building conversational agents},
 type = {misc},
 year = {2019},
 source = {arXiv},
 id = {3322a4de-833a-3fd0-a6ad-ad82e005daf7},
 created = {2020-10-27T23:59:00.000Z},
 file_attached = {false},
 profile_id = {d7d2e6da-aa5b-3ab3-b3f2-a5350adf574a},
 last_modified = {2020-10-30T08:03:31.996Z},
 read = {false},
 starred = {false},
 authored = {true},
 confirmed = {false},
 hidden = {false},
 private_publication = {false},
 abstract = {Copyright © 2019, arXiv, All rights reserved. We have recently seen the emergence of several publicly available Natural Language Understanding (NLU) toolkits, which map user utterances to structured, but more abstract, Dialogue Act (DA) or Intent specifications, while making this process accessible to the lay developer. In this paper, we present the first wide coverage evaluation and comparison of some of the most popular NLU services, on a large, multi-domain (21 domains) dataset of 25K user utterances that we have collected and annotated with Intent and Entity Type specifications and which will be released as part of this submission.1 The results show that on Intent classification Watson significantly outperforms the other platforms, namely, Dialogflow, LUIS and Rasa; though these also perform well. Interestingly, on Entity Type recognition, Watson performs significantly worse due to its low Precision2. Again, Dialogflow, LUIS and Rasa perform well on this task.},
 bibtype = {misc},
 author = {Liu, X. and Eshghi, A. and Swietojanski, P. and Rieser, V.}
}

{"_id":"k8gj5zJFjE4tSrKys","bibbaseid":"liu-eshghi-swietojanski-rieser-benchmarkingnaturallanguageunderstandingservicesforbuildingconversationalagents-2019","author_short":["Liu, X.","Eshghi, A.","Swietojanski, P.","Rieser, V."],"bibdata":{"title":"Benchmarking natural language understanding services for building conversational agents","type":"misc","year":"2019","source":"arXiv","id":"3322a4de-833a-3fd0-a6ad-ad82e005daf7","created":"2020-10-27T23:59:00.000Z","file_attached":false,"profile_id":"d7d2e6da-aa5b-3ab3-b3f2-a5350adf574a","last_modified":"2020-10-30T08:03:31.996Z","read":false,"starred":false,"authored":"true","confirmed":false,"hidden":false,"private_publication":false,"abstract":"Copyright © 2019, arXiv, All rights reserved. We have recently seen the emergence of several publicly available Natural Language Understanding (NLU) toolkits, which map user utterances to structured, but more abstract, Dialogue Act (DA) or Intent specifications, while making this process accessible to the lay developer. In this paper, we present the first wide coverage evaluation and comparison of some of the most popular NLU services, on a large, multi-domain (21 domains) dataset of 25K user utterances that we have collected and annotated with Intent and Entity Type specifications and which will be released as part of this submission.1 The results show that on Intent classification Watson significantly outperforms the other platforms, namely, Dialogflow, LUIS and Rasa; though these also perform well. Interestingly, on Entity Type recognition, Watson performs significantly worse due to its low Precision2. Again, Dialogflow, LUIS and Rasa perform well on this task.","bibtype":"misc","author":"Liu, X. and Eshghi, A. and Swietojanski, P. and Rieser, V.","bibtex":"@misc{\n title = {Benchmarking natural language understanding services for building conversational agents},\n type = {misc},\n year = {2019},\n source = {arXiv},\n id = {3322a4de-833a-3fd0-a6ad-ad82e005daf7},\n created = {2020-10-27T23:59:00.000Z},\n file_attached = {false},\n profile_id = {d7d2e6da-aa5b-3ab3-b3f2-a5350adf574a},\n last_modified = {2020-10-30T08:03:31.996Z},\n read = {false},\n starred = {false},\n authored = {true},\n confirmed = {false},\n hidden = {false},\n private_publication = {false},\n abstract = {Copyright © 2019, arXiv, All rights reserved. We have recently seen the emergence of several publicly available Natural Language Understanding (NLU) toolkits, which map user utterances to structured, but more abstract, Dialogue Act (DA) or Intent specifications, while making this process accessible to the lay developer. In this paper, we present the first wide coverage evaluation and comparison of some of the most popular NLU services, on a large, multi-domain (21 domains) dataset of 25K user utterances that we have collected and annotated with Intent and Entity Type specifications and which will be released as part of this submission.1 The results show that on Intent classification Watson significantly outperforms the other platforms, namely, Dialogflow, LUIS and Rasa; though these also perform well. Interestingly, on Entity Type recognition, Watson performs significantly worse due to its low Precision2. Again, Dialogflow, LUIS and Rasa perform well on this task.},\n bibtype = {misc},\n author = {Liu, X. and Eshghi, A. and Swietojanski, P. and Rieser, V.}\n}","author_short":["Liu, X.","Eshghi, A.","Swietojanski, P.","Rieser, V."],"biburl":"https://bibbase.org/service/mendeley/d7d2e6da-aa5b-3ab3-b3f2-a5350adf574a","bibbaseid":"liu-eshghi-swietojanski-rieser-benchmarkingnaturallanguageunderstandingservicesforbuildingconversationalagents-2019","role":"author","urls":{},"metadata":{"authorlinks":{}}},"bibtype":"misc","biburl":"https://bibbase.org/service/mendeley/d7d2e6da-aa5b-3ab3-b3f2-a5350adf574a","dataSources":["jurZeGzSpYdkQ8rm4","ya2CyA73rpZseyrZ8","TcQYToyGTfqDApS68","BEu9hLf9unY5A9Pwe","gQ3XnmcCc7p5JYvBa","pB6WyiWKHxyFpemKA","2252seNhipfTmjEBQ"],"keywords":[],"search_terms":["benchmarking","natural","language","understanding","services","building","conversational","agents","liu","eshghi","swietojanski","rieser"],"title":"Benchmarking natural language understanding services for building conversational agents","year":2019}