Cyberspace, data analytics, and policing
In: A Chapman & Hall book
PrefaceList of FiguresList of TablesIntroductionCyberspace2.1 What is cyberspace?2.2 The impact of cyberspace2.3 Identity and authentication2.4 Encryption2.5 Crime is changing2.6 Policing is changingNew opportunities for criminality3.1 Unprecedented access to information3.2 Crimes directed against cyberspace3.2.1 Malware3.2.2 Crimes of destruction3.2.3 Monetized cybercrimes3.2.4 Data theft crimes3.2.5 Secondary markets3.3 Crimes that rely on cyberspace3.3.1 Spam, scams, and cons3.3.2 Financial crime3.3.3 Online shopping3.3.4 Crimes against children3.4 Crimes done differently because of cyberspace3.4.1 Disseminating hatred3.4.2 Selling drugs3.4.3 Stalking and crime preparation3.4.4 Digital vigilantes3.5 Money laundering3.5.1 Cash3.5.2 The financial system3.5.3 International money laundering3.5.4 Cryptocurrencies3.6 Overlap with violent extremismNew ways for criminals to interact4.1 Criminal collaboration4.2 Planning together4.3 Information sharing4.3.1 Sharing techniques4.3.2 Sharing resources4.3.3 Sharing vulnerabilities4.4 International interactionsData analytics makes criminals easier to find5.1 Understanding by deduction5.2 Understanding by induction5.3 Subverting data analytics5.4 Intelligence-led policing5.5 Hot spot policing5.5.1 Place5.5.2 Time5.5.3 Weather5.5.4 People involved5.5.5 Social network position5.6 Exploiting skewed distributionsData collection6.1 Ways to collect data6.2 Types of data collected6.2.1 Focused data6.2.2 Large volume data6.2.3 Incident data6.2.4 Spatial data6.2.5 Temporal data6.2.6 Non-crime data6.2.7 Data fusion6.2.8 Protecting data collected by law enforcement6.3 Issues around data collection6.3.1 Suspicion6.3.2 Wholesale data collection6.3.3 Privacy6.3.4 Racism and other -isms6.3.5 Errors6.3.6 Bias6.3.7 Sabotaging data collection6.3.8 Getting better data by sharingTechniques for data analytics7.1 Clustering7.2 Prediction7.3 Meta issues in prediction7.3.1 Classification versus regression7.3.2 Problems with the data7.3.3 Why did the model make this prediction?7.3.4 How good is this model?7.3.5 Selecting attributes7.3.6 Making predictions in stages7.3.7 Bagging and boosting7.3.8 Anomaly detection7.3.9 Ranking7.3.10 Should I make a prediction at all?7.4 Prediction techniques7.4.1 Counting techniques7.4.2 Optimization techniques7.4.3 Other ensembles7.5 Social network analysis7.6 Natural language analytics7.7 Making data analytics available7.8 Demonstrating complianceCase studies8.1 Predicting crime rates8.2 Clustering RMS data8.3 Geographical distribution patterns8.4 Risk of gun violence8.5 Copresence networks8.6 Criminal networks with a purpose8.7 Analyzing online posts8.7.1 Detecting abusive language8.7.2 Detecting intent8.7.3 Deception8.7.4 Detecting fraud in text8.7.5 Detecting sellers in dark-web marketplaces8.8 Behavior ⁰́₃ detecting fraud from mouse movements8.9 Understanding drug trafficking pathwaysLaw enforcement can use interaction too9.1 Structured interaction through transnational organizations9.2 Divisions within countries9.3 Sharing of information about crimes9.4 Sharing of data9.5 Sharing models9.6 International issuesSummaryIndex