A Framework for Domain-Specific Natural Language Information Brokerage

Abstract

Service providers — from public institutions to primary care facilities — need to constantly attend to clients’ inquiries to provide useful information and directive guidelines. Ensuring high quality service is challenging as it not only demands detailed domain-specific knowledge, but also the ability to quickly understand the clients’ issues through their diverse — and often casual — descriptions. This paper aims to provide a framework for the development of an automated information broker agent who performs the task of a helper. The main task of the agent is to interact with the client and direct them to obtain further services that cater their personalized need. To do so, the agent should accomplish a sequence of tasks that include natural language inquiry, knowledge gathering, reasoning, and giving feedback; in this way, it simulates a human helper to engage in interaction with the client. The framework combines a question-answering reasoning mechanism while utilizing domain-specific knowledge base. When the users cannot describe clearly their needs, the system tries to narrow down the possibilities by an iterative question-answering process, until it eventually identifies the target. In realizing our framework, we make a proof-of-concept project, Mandy, a primary care chatbot system created to assist healthcare staffs by automating the patient intake process. We describe in detail the system functionalities and design of the system, and evaluate our proof-of-concept on benchmark case studies.

Keywords

Question and answer system / chatbot / automated information broker / iterative inquiry / language processing / AI and healthcare

References

[1] AhmadF., Hogg-JohnsonS., StewartD. E., SkinnerH. A., GlazierR. H., LevinsonW.. Computer-assisted screening for intimate partner violence and controla randomized trial. Annals of internal medicine, 2009, 151(2): 93-102

[2] ArazO. M.. Integrating complex system dynamics of pandemic influenza with a multi-criteria decision making model for evaluating public health strategies. Journal of Systems Science and Systems Engineering, 2013, 22: 319-339

[3] BakkerM., TsuiK.L.. Dynamic resource allocation for efficient patient scheduling: A data-driven approach. Journal of Systems Science and Systems Engineering, 2017, 26(4): 448-462

[4] BendigA. W.. Twenty questions: an information analysis. Journal of Experimental Psychology, 1953, 46(5): 345

[5] BengioY., DucharmeR., VincentP., JauvinC.. A neural probabilistic language model. Journal of machine learning research, 3(Feb), 20031137-1155

[6] BernsteinS. L., AronskyD., DusejaR., EpsteinS., HandelD., HwangU., McCarthyM. J., McConnellK., PinesJ. M., RathlevN., et al.. The effect of emergency department crowding on clinically oriented outcomes. Academic Emergency Medicine, 2009, 16(1): 1-10

[7] BickmoreT., GiorginoT.. Health dialog systems for patients and consumers. Journal of biomedical informatics, 2006, 39(5): 556-571

[8] BickmoreT. W., PfeiferL. M., ByronD., ForsytheS., HenaultL. E., JackB. W., SillimanR., Paasche-OrlowM. K.. Usability of conversational agents by patients with inadequate health literacy: evidence from two clinical trials. Journal of health communication, 2010197-210

[9] BlackL.A., McTearM., BlackN., HarperR., LemonM.. Appraisal of a conversational artefact and its utility in remote patient monitoring. In Computer-Based Medical Systems, 2005. Proceedings. 18th IEEE Symposium on, 2005506-508

[10] BobrowD. G.A question-answering system for high school algebra word problems, 1964591-614

[11] BodenreiderO.The unified medical language system (umls): integrating biomedical terminology, 2004

[12] CaleyM., SidhuK.. Estimating the future healthcare costs of an aging population in the uk: expansion of morbidity and the need for preventative care. Journal of Public Health, 2011, 33(1): 117-122

[13] ChenP., CheungY., LeeV. C., HartA.. Knowledge sharing via informal communities in a globally distributed organization. In International Symposium on Knowledge and Systems Sciences, pages, 201630-43

[14] ClanceyW. J., LetsingerR.NEOMYCIN: Reconfiguring a rule-based expert system for application to teaching, 1982

[15] DangT.T., HoT.B.. Sequence-based measure for assessing drug-side effect causal relation from electronic medical records. In International Symposium on Knowledge and Systems Sciences, 201753-65

[16] De VineL., ZucconG., KoopmanB., SitbonL., BruzaP.. Medical semantic similarity with a neural language model. In Proceedings of the 23rd ACM international conference on conference on information and knowledge management, 20141819-1822

[17] DelichatsiosH. K., FriedmanR. H., GlanzK., TennstedtS., SmigelskiC., PintoB. M., KelleyH., GillmanM. W.. Randomized trial of a “talking computer” to improve adults’ eating habits. American Journal of Health Promotion, 2001, 15(4): 215-224

[18] Di SommaS., PaladinoL., VaughanL., LalleI., MagriniL., MagnantiM.. Overcrowding in emergency department: an international issue. Internal and emergency medicine, 2015, 10(2): 171-175

[19] DobbieG., RossK.. Precision driven health: A new zealand research partnership. International Journal of Integrated Care, 2017

[20] EdgrenL.. Health consumer diversity and its implications. Journal of Systems Science and Systems Engineering, 2006, 15(1): 34-47

[21] FarzanfarR., FrishkopfS., MigneaultJ., FriedmanR.. Telephone-linked care for physical activity: a qualitative evaluation of the use patterns of an information technology program for patients. Journal of Biomedical Informatics, 2005, 38(3): 220-228

[22] FerrucciD., BrownE., Chu-CarrollJ., FanJ., GondekD., KalyanpurA. A., LallyA., MurdockJ. W., NybergE., PragerJ., et al.. Building watson: An overview of the deepqa project. AI magazine, 2010, 31(3): 59-79

[23] GreenB. F.Jr, WolfA. K., ChomskyC., LaugheryK.Baseball: an automatic question-answerer, 1961219-224

[24] HerzogO., RollingerC.R.Text understanding in LILOG: integrating computational linguistics and artificial intelligence: final report on the IBM Germany LILOG-Project, 1991

[25] HighR.The era of cognitive systems: An inside look at ibm watson and how it works, 2012

[26] HubalR. C., DayR. S.. Informed consent procedures: An experimental test using a virtual character in a dialog systems training application. Journal of Biomedical Informatics, 2006, 39(5): 532-540

[27] HuntD. L., HaynesR. B., HannaS. E., SmithK.. Effects of computer-based clinical decision support systems on physician performance and patient outcomes: a systematic review. Jama, 1998, 280(15): 1339-1346

[28] JiangB., TangJ., YanC.. Outpatient capacity allocation considering adding capacity to match high patient demand. Journal of Systems Science and Systems Engineering, 2017, 26(4): 487-516

[29] JonesP., ChalmersL., WellsS., AmeratungaS., CarswellP., AshtonT., CurtisE., ReidP., StewartJ., HarperA., et al.. Implementing performance improvement in new zealand emergency departments: the six hour time target policy national research project protocol. BMC health services research, 2012, 12(1): 45

[30] KhouryM. J., IoannidisJ. P.. Big data meets public health. Science, 2014, 346(6213): 1054-1055

[31] LipkinM., QuillT. E., NapodanoR. J.. The medical interview: a core curriculum for residencies in internal medicine. Annals of Internal Medicine, 1984, 100(2): 277-284

[32] LiuY., GeT., MathewsK., JiH., McGuinnessD.. Exploiting task-oriented resources to learn word embeddings for clinical abbreviation expansion. Proceedings of BioNLP 15, 201592-97

[33] Martínez-PérezB., de la Torre-DíezI., López-CoronadoM., Sainz-De-AbajoB., RoblesM., García-GómezJ. M.. Mobile clinical decision support systems and applications: a literature and commercial review. Journal of medical systems, 2014, 38(1): 4

[34] McFillenJ. M., O’NeilD. A., BalzerW. K., VarneyG. H.. Organizational diagnosis: An evidence-based approach. Journal of Change Management, 2013, 13(2): 223-246

[35] MikolovT., ChenK., CorradoG., DeanJ.Efficient estimation of word representations in vector space, 2013

[36] MikolovT., SutskeverI., ChenK., CorradoG. S., DeanJ.. Distributed representations of words and phrases and their compositionality. In Advances in neural information processing systems, 20133111-3119

[37] MillerR. A. P., JrH. E., MyersJ. D.. Internist-i, an experimental computer-based diagnostic consultant for general internal medicine. New England Journal of Medicine, 1982, 307(8): 468-476

[38] MoskvinaA., LiuJ.. How to build your network? a structural analysis. In Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, 20162597-2603

[39] MoskvinaA., LiuJ.. Integrating networks of equipotent nodes. In International Conference on Computational Social Networks, 201639-50

[40] NiL., LuC., LiuN., LiuJ.. Mandy: Towards a smart primary care chatbot application. In International Symposium on Knowledge and Systems Sciences, 201738-52

[41] NikfarjamA., SarkerA., O’ConnorK., GinnR., GonzalezG.. Pharmacovigilance from social media: mining adverse drug reaction mentions using sequence labeling with word embedding cluster features. Journal of the American Medical Informatics Association, 2015, 22(3): 671-681

[42] PathakJ., SolbrigH. R., BuntrockJ. D., JohnsonT. M., ChuteC. G.. Lexgrid: a framework for representing, storing, and querying biomedical terminologies from simple to sublime. Journal of the American Medical Informatics Association, 2009, 16(3): 305-315

[43] PechsiriC., SukharomanaR.. Comparative study of using word co-occurrence to extract disease symptoms from web documents. In International Symposium on Knowledge and Systems Sciences, 201791-100

[44] RamelsonH. Z., FriedmanR. H., OckeneJ. K.. An automated telephone-based smoking cessation education and counseling system. Patient Education and Counseling, 1999, 36(2): 131-144

[45] RealdiG., PreviatoL., VitturiN.. Selection of diagnostic tests for clinical decision making and translation to a problem oriented medical record. Clinica Chimica Acta, 2008, 393(1): 37-43

[46] RichardsonD. B.. Increase in patient mortality at 10 days associated with emergency department overcrowding. The Medical Journal of Australia, 2006, 184(5): 213-216

[47] SayginA. P., CicekliI., AkmanV.. Turing test: 50 years later. Minds and Machines, 2000, 10(4): 463-518

[48] SternS., CifuA., AltkornD.Symptom to Diagnosis An Evidence Based Guide. McGraw Hill Professional, 2014

[49] TaberJ. M., LeyvaB., PersoskieA.. Why do people avoid medical care? a qualitative study using national data. Journal of general internal medicine, 2015, 30(3): 290-297

[50] TienJ. M., BergD.. A case for service systems engineering. Journal of systems science and systems engineering, 2003, 12(1): 13-38

[51] TienJ. M., Goldschmidt-ClermontP. J.. Healthcare: A complex service system. Journal of Systems Science and Systems Engineering, 2009, 18(3): 257-282

[52] VictorL. Y., BuchananB. G., ShortliffeE. H., WraithS. M., DavisR., ScottA. C., CohenS. N.. Evaluating the performance of a computer-based consultant. Computer programs in biomedicine, 1979, 9(1): 95-102

[53] WangC., TangX.. The online debate networks analysis: A case study of debates at tianya forum. In International Symposium on Knowledge and Systems Sciences, 2016140-150

[54] WarrenJ. R.. Better, more cost-effective intake interviews. IEEE Intelligent Systems and their Applications, 1998, 13(1): 40-48

[55] WeizenbaumJ.. Eliza—a computer program for the study of natural language communication between man and machine. Communications of the ACM, 1966, 9(1): 36-45

[56] WilenskyR., ChinD. N., LuriaM., MartinJ., MayfieldJ., WuD.. The berkeley unix consultant project. Computational Linguistics, 1988, 14(4): 35-84

[57] WinogradT.Procedures as a representation for data in a computer program for understanding natural language, 1971

[58] WoodsW. A., KaplanR.. Lunar rocks in natural english: Explorations in natural language question answering. Linguistic structures processing, 1977, 5: 521-569

[59] XuJ., ZhangY., XuH., et al.. Clinical abbreviation disambiguation using neural word embeddings. Proceedings of BioNLP 15, 2015171-176

[60] YanB., ChenY., LiuJ.. Dynamic relationship building: exploitation versus exploration on a social network. In International Conference on Web Information Systems Engineering, 201775-90

[61] YanB., LiuY., LiuJ., CaiY., SuH., ZhengH.. From the periphery to the center: Information brokerage in an evolving network. In Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, IJCAI 2018, 20183912-3918

[62] ZhangW., JiangY., YoshidaT.. Deep context identification of deceptive reviews using word vectors. In International Symposium on Knowledge and Systems Sciences, 2016213-224

RIGHTS & PERMISSIONS

Systems Engineering Society of China and Springer-Verlag GmbH Germany, part of Springer Nature