Cloud computing in telecom space Part1

Cloud computing in telecom space:
Cloud Computing is a buzzword in the telecoms industry.By moving into the Cloud Computing arena, and beyond their role as mere connectivity providers, operators were hoping that by applying cloud services mainly like Iaas(Infrastructure as Service),Paas(Platform as a Service) and SaaS(Software as a Service) in telecom spaces they are going to get huge benifit. A Major issue in cloud computing,especially while using public clouds with Privacy,Security and with Standard Complaince.So Industry is going through a common framework where all issues can be addreseed in cloud computing space.
Major benifits of cloud computing in telecom space were
Competition, cost pressures, and the demand for services and applications anytime, anywhere, and on any device are forcing telecom service providers to consider alternative model and deleiver IT services demanded by the customers.
Read more...

Fraud Management System Part5(Investigation or Analysis Techniques)

Investigation or Analysis Techniques for Fraud Management :
• Rule based Analysis
• Data Mining (scoring module)
• Profiling
• Fingerprinting
• Link Analysis
Rule-based Analysis
• Used to customize how Alarm Analysis performs its work, based on operator unique policies and procedures
• Rule Conditions
− Existence of a specified severity level of a particular Alarm Count or Threshold Alarm
• Rule Results
− Create a case of a certain fraud type
− Take/Recommend an action
− Route a case on a specific worklist
Data Mining Analysis
• Activity parallel to rule analysis
• For each new alarm, the customer will be evaluated using existing scoring models
• SPSS Clementine provide the mechanism to extend and adapt the base models
• The resulting score will be taken into account by the workflow engine (case creation/routing)
New Case Redirection
Case Investigation --Supporting Information
• Displays information about the Case, including a list of Alarms detected and calls
• View Alarm or Call details
• Also, view:
− Entity information (general, contact, usage, profile)
− Actions (recommended by system, as well as analyst)
− System Findings
− Notes (entered by analyst, past or present)
− Additional investigation data (from other systems)
Case Creation Policies
>Correlates Alarms
>Generates case with confidence level
>Provides recommended actions

Alarm AnalysisCase Resolution
Case managementCase List summary
• Case Status
− Unseen not yet looked at by analyst
− Open an analyst is working on it
− On Watch analyst has reviewed and is not sure if fraud, watching for further evidence
− Alarmed case “on watch” by analyst, and new alarms have occurred for the customer
• Sort on
− Fraud Type detected, Service #, Confidence Level, Date Case created or modified, # of Alarms
• Select a Case to work on
Case Worklists
• Worklists define which analysts can see or work on different cases
− Individual Worklists
− Shared Worklists
• Allow fraud manager to organize FMS results to fit into their organizational structure
− subgroups or individuals that specialize in different networks or types of fraud
− cases for VIPs or sensitive subscribers
Automatic ActionsAllows Automatic Response To Fraud
• Service Providers’ Policies and Procedures
• Logical vs. Physical Actions
• Interface to external systems
• Invokes a shell script
− Shell script parameters
• equipment identifier
• service identifier
• etc.
Interactive ActionsAllow Manual Control
• Allow customers to access Web-Enabled applications through the FMS client
• System integrators will be able to:
− create custom buttons
− assign text and tool tips
− specify bitmaps and tie them to different customer applications
• FMS will be able to open a window displaying the contents of the URL associated with the button
Link Analysis & Call TracerInvestigation Analysis for subscription fraud
• Allows analyst to identify
− when two or more cases had calls to a common number
− when multiple fraudulent subscribers were called by a common number
• Link exploration
− Displaying in graphical form all numbers called by the target
− Reversing direction, and displaying in graphical form all numbers that called the target
− Iterate through multiple levels of linkage
Link Analysis
Case Archive
• Store information for resolved cases
− whether actual fraud or not
− fraud types, alarms, calls, actions
• The purpose is three-fold:
− cases can be retrieved and shown in the user interface
− input to data mining case analysis
− source of data carriers may use for their own analysis
FMS user access
• Four types of Users
− Case Manager fraud analyst
− Knowledge Manager implementing carrier’s fraud management policies
− System Manager responsible for overall system stability and proper performance
− Security Manager user management and ensuring proper use by the users
• Other privileges
− white list, add to black list, link analysis, custom toolbar, call archive query
• Access Buttons
− only enabled if authorized
FMS User Interface
• Explorer-type GUI supports pane-based editing
− Information organized in hierarchal view
− Easy navigation to make user more efficient
• Graphs
− New graphs combine thresholds, usage etc
• Print function on windows
• New editing capabilities for black and white lists
• User- and time-stamped Notes
Reports
− Only authorized users can create each class of report
− Case, Knowledge, and System Management
− HTML generated reports
− Custom reports can be developed for you
− In addition, carrier can use any Oracle compliant reporting tool to customize other types of reports (e.g. Business Objects)
Read more...