{"title":"Natural language interface to a corporate cash management system","authors":"M. D. Harris","doi":"10.1109/AIAWS.1991.236578","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236578","url":null,"abstract":"The development is described of a natural language interface to the corporate cash management software of a major bank. The author presents an overview of natural language processing technology, discusses the project's advantages and limitations, and suggests other applications for NLP in banking systems.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123474037","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"TRADER: an expert system for trading commodities futures","authors":"C. Brown, I. Holland, R. Mesch","doi":"10.1109/AIAWS.1991.236567","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236567","url":null,"abstract":"An expert system is described, written in PROLOG, for trading commodities futures. The system uses a blackboard architecture to develop a set of confidence factors for buying and selling futures. It also has an elaborate user interface, which displays the technical studies that its decisions are based on in graphical form and provides feedback to the user on its decisions. This interface played a major role in the knowledge acquisition phase of the project. The program is a real-time system using tick data provided by Tick Data Inc. of Chicago, IL. It has been extensively tested on historical data with encouraging results; the authors are refining the confidence factor computation and adding a training aspect to the program.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130498679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Management of uncertainty for intelligent financial systems","authors":"E. Cortes-Rello, F. Golshani","doi":"10.1109/AIAWS.1991.236595","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236595","url":null,"abstract":"A fundamental function of financial intermediaries is to assess and assume risk. Financial intermediaries have to deal with risks such as credit risks, liquidity risks, interest rate changes and exchange rate changes. Some of these risks can be dealt with using techniques such as hedging, but there is always some subjective judgement involved. The relevance of the Dempster-Shafer theory of evidence to financial problems that involve subjective judgement is studied. Three specific examples are analyzed: forecasting, cross-hedging and financial group decision making.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"65 11","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120933487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"A knowledge-based approach to solving hedge design problems","authors":"M. Benaroch, V. Dhar","doi":"10.1109/AIAWS.1991.236558","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236558","url":null,"abstract":"Problems such as the explosive number of hedging alternatives that is constantly growing, the quick decisions risk managers need to make in response to the speed in which information flows, and the lack of appropriate computerized support in the early phases of the hedge design process, suggest that there is ample scope for risk managers to make suboptimal decisions. The authors formulate hedge design as a multi-objective optimization problem. This optimization problem involves several complexities with which existing quantitative solution techniques cannot deal. They also present a knowledge-based system called INTELLIGENT-HEDGER, developed to solve the hedge design problem. This system uses an object-centered representation that captures risk managers' deep domain knowledge, and facilitates emulation of first-principles reasoning processes risk managers use to make decisions and explain them.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128543093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Building a banking system specification using machine learning","authors":"J. Genest","doi":"10.1109/AIAWS.1991.236591","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236591","url":null,"abstract":"Transforming user requirements into software specification is a complex and demanding task. Artificial intelligence methods such as machine learning (ML) can assist in the software specification process by providing support to system designers. This paper presents an approach based on explanation-based learning (EBL), a ML technique in which a concept is learned by building an explanation. The approach is presented in the context of the system LISE (Learning in Software Engineering). LISE converts a user requirement for a software module into an operational module definition using EBL with an incomplete theory. An example where LISE is used to build the specification of a banking system is illustrated.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128676760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Neural networks and the mathematics of chaos-an investigation of these methodologies as accurate predictors of corporate bankruptcy","authors":"D. Cadden","doi":"10.1109/AIAWS.1991.236574","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236574","url":null,"abstract":"Research is given which investigates the use of neural networks as a classification tool to predict corporate bankruptcy. It also examines if a branch of chaos theory could be used as a forecasting tool. It compares neural networks with an alternative corporate bankruptcy prediction model Altman's Z-score model (1977) on a small data sample and illustrates that neural networks can yield superior results.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"208 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115903559","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Margin credit evaluation system","authors":"R. Beshinske, S. Spence, R. Nigam","doi":"10.1109/AIAWS.1991.236561","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236561","url":null,"abstract":"The Management Science Group developed an expert system for the Credit Administration Department at Merrill Lynch to assist them with their primary responsibility of setting credit limits on margin accounts. Every account with a debit balance over $100000 is delivered daily to the system, appraised, and prioritized by its risk, using a combination of analytic and heuristic reasoning to evaluate its equity and fixed income positions on the basis of quality, marketability, and overall portfolio diversity. The system warns the credit analyst about problems in the account, and gives advice about whether the current debit can be supported and what the credit limits should be. The account may also be analyzed on an interactive basis, and what-if scenarios may be run to explore the effects of modifying the portfolio. Merrill Lynch has made daily operational use of the system for over two years and uses it to oversee more than two billion dollars of margin credit.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129977990","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Program traffic locator-a knowledge based expert system","authors":"S. R. Nair","doi":"10.1109/AIAWS.1991.236559","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236559","url":null,"abstract":"A program basket contains a specified set of stocks in a certain proportion. It is a difficult problem to recognize a program basket when the orders comprising the basket are interspersed with the large number of individual retail orders arriving at the stock exchange. A knowledge based expert system was designed to identify and 'learn' the basket properties so that it can be used to identify baskets using approximate matching.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133492431","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"From technology push to market pull: expert systems in the front office","authors":"S. Davis","doi":"10.1109/AIAWS.1991.236598","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236598","url":null,"abstract":"Takes a pragmatic view as to why expert systems have not been the success they were anticipated to be in the trading rooms. The author invites expert systems developers to take a more realistic view of what users are demanding and encourages them to let go of their own aspirations for the technology. The key to success is routine and sometimes dull applications that help solve real business problems. Such applications are likely to require the ability to reason in real time and about how events are changing over time. Combined with improved systems integration, the age of simple but useful expert systems for the front office could soon be upon us.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116613788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Adding value to financial news by computer","authors":"P. Hayes, S. P. Weinstein","doi":"10.1109/AIAWS.1991.236581","DOIUrl":"https://doi.org/10.1109/AIAWS.1991.236581","url":null,"abstract":"Recent advances in software permit applications where computers can intelligently process text based on its content. Since so much financial information is available first (and sometimes only) in textual form, these developments have great significant for Wall Street. The authors describe a deployed application that enhances access to financial news stories by categorizing them (as corporate acquisitions, earning reports, etc.) on the basis of their content. They go on to describe other applications possible with more advanced forms of this technology, involving extraction of key facts from certain types of stories. These applications include summarization, creation of databases from news feeds, translation into other languages, and even program trading based on news. An application in this class is scheduled for deployment in 1991.<<ETX>>","PeriodicalId":171318,"journal":{"name":"Proceedings First International Conference on Artificial Intelligence Applications on Wall Street","volume":"66 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1991-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115722339","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}