Object-Relational CDR Analysis for High-Scale Low-Signal Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Analysis of high-scale low-signal (HSLS) data, such as call data records (CDRs), is laborious and time-consuming due to the unstructured and voluminous nature of the raw data, often provided from multiple sources with varying formats.
Innovation Solution
The implementation of an object-relational model to organize CDRs into a database, allowing for efficient storage and rapid access through object types, properties, and relationships, followed by query-based filtering and analysis to generate insightful subsets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If raw CDR data from multiple sources is analyzed directly, then comprehensive investigation coverage is achieved, but analysis time and computational resources increase significantly
Solution Approach 1:
The system performs preliminary actions by pre-processing raw CDR data from multiple sources and organizing it into an object-relational database model before analysis is requested. This includes standardizing data formats, creating normalized schemas, and establishing relationships between data entities in advance, so that when investigation queries are executed, the data is already structured and ready for rapid retrieval and analysis.
2Ease of manufacture
If raw CDR data is stored in unstructured format, then storage simplicity is maintained, but data retrieval and analysis efficiency decrease
Solution Approach 1:
The system segments raw CDR data into distinct object types (e.g., call records, subscriber information, location data, device details) and stores them in separate relational tables within the database. Each object type has its own structured schema with defined properties and relationships, allowing efficient querying and retrieval of specific data elements without processing entire raw datasets.
3Loss of information
If comprehensive CDR datasets are processed, then complete investigation information is obtained, but processing complexity and resource requirements increase
Solution Approach 1:
The system extracts only the relevant data elements needed for investigations from comprehensive CDR datasets and stores them in the object-relational database model. By taking out and isolating specific object types and their properties (such as call details, location information, device identifiers), the system maintains complete investigation information while reducing processing complexity through focused, structured storage.
Data Source
AI summary
An apparatus, computer-implemented method and computer program are disclosed for organization and analysis of high-scale low-signal (HSLS) data. The method may comprise providing one or more raw datasets representing CDRs and processing the raw datasets to generate an object-relational model comprising a plurality of objects, each object having an object type and properties of the respective object type. The method may also comprise receiving at least a first query comprising, for at least one object type of the object-relational model, at least a first property relating to a first investigation and applying the at least first query to the stored object-relational model to generate a first subset of the CDRs. Responsive to selection of an analysis command, the method may comprise performing a predetermined type of analysis associated with the analysis command on the first subset of CDRs to produce a resultant subset of the CDRs.