A System For Taxi Crime Prevention And Help

There is no standard definition of artificial intelligence.Generally speaking,if a computer system can do whatever that requires humans' intelligence if humans do it,the computer system has artificial intelligence.The purpose of human creation of artificial intelligence systems is to provide human life better via using such systems.More specifically,artificial intelligence has been widely applied in various fields that need human intelligence,such as law,medicine,finance,and e-commerce.Among these fields,the law is one of the most important domains of artificial intelligence applications currently.This is because artificial intelligence systems can operate in accordance with human-designed algorithms,rules,and templates,thereby freeing legal workers from tedious and repetitive tasks and allowing them to focus more on tasks that require creative and interpersonal communication.Thus,in this thesis,we first survey the research state-of-art and development trend of the application of artificial intelligence in the legal field and the legislation of artificial intelligence systems.Artificial intelligence can be applied to all aspects of the legal domain,including leg-islation,knowledge and compliance(legal information retrieval,legal document generation and review),law enforcement(online evidence searching,crime prevention,legal reasoning,and on-line dispute resolution).In particular,key artificial intelligence technologies of legal information retrieval include similarity measure methods,multi-intelligent agents,natural language processing technology,text mining technology,and multi-attribute decision-making methods.And the ap-plications of artificial intelligence in aspect of evidence collection and crime prevention include online evidence searching,using online evidence to predict crime,cybercrime detection,security game,suicide monitoring and warning,and Uber and Didi taxi crime prevention.Regarding the legislation of artificial intelligence systems,the current hot issues are how to set traffic laws for autonomous vehicles,the division of responsibilities for accidents,and the automatic decision-making when any moral dilemmas occur.Replacing some jobs in the judicature field with artificial intelligence systems has become a reality and replacing more jobs will become a trend,but the widespread application of artificial intelligence has also created many new jobs.The difficulty in applying artificial intelligence to the law may be that there is no uniform answer to legal issues in practice.For example,it will be affected by money,emotions,and various practical situations,and it is not yet possible to truly understand the real world(and experienced lawyers may not be able to do so).This lack of standard experience has become one of the main factors hindering the development of artificial intelligence justice.If intelligent systems can really think like humans,the development of artificial intelligence will reach a higher peak.In the application domain of artificial intelligence in law,in this thesis,we uses artificial intelligence to develop a crime prevention system.The motivation that we develop the system is as follows.With the changes of the times and the advancement of science,there are more and more cars around us.Cars bring great convenience to our daily life.For example,Didi taxi in China can be on-call at any time and anywhere,cheaper and convenient.However,in recent years,there have been many incidents of female passengers being killed by Didi taxi drivers.As a result,females are so afraid of taking a taxi out.To address this issue,in this thesis,we develop an artificial intelligence system using voice input to identify crime tendency.The system we developed uses the keyword matching technology of speech input,more specifically,we first build up the keyword library of criminal trends based on similar cases re-ported in the news.Our system uses these keywords to identify the serious degrees of a criminala's criminal tendency:(1)criminal rhetoric has been strongly expressed,and criminal behaviour is in progress;(2)the criminal rhetoric has been clearly expressed,so the criminal act is about to pro-ceed;(3)the desire to commit crimes has been implicitly expressed,and further action is likely.Then the system checks up the keywords in the incoming record of a user.The speech recognition part is developed in the Android Studio environment based on the Google Android platform and iFLYTEK voice dictation SDK.Once the system identifies a keyword in the record,it will automat-ically warn the driver and meanwhile,send the relative a distress signal and relevant information.Our system also saves the audio record of the driver words and the text recognised in the local device and send them relatives and friends as evidence of the crime.The system elicits location information where the taxi is through the Android platform using the Gaode Map SDK,which uses GPS+base station+WiFi mixed mode for positioning.Our system updates the location information per 3 seconds.Under the Android operating system,we have implemented our speech recognition crime detection and distress system based on keyword matching and tested it with real conversation scenarios similar to those of Didi taxi passengers killed.The quality evaluation of our system in detecting crime tendency shows that the system has practical significance in crime prevention.Compared with the case of evidence search which has been confirmed before the facts of crime,or the case of crime analysis by detecting abnormal behaviours on the Internet,the advantage of our detection and distress system of speech recognition criminal tendency is that it can carry out real-time evidence searching and sending out distress signals.Therefore,our system is very useful in real life.