REAL TIME ANALYTICS PIPELINE FOR TRANSACTION ABNORMALITY DETECTION
A real time analytics pipeline needed to be built to detect any abnormal transactions between customers of a large telecom services company. The pipeline had to be deployed on the cloud.
Our consultants used Amazon kinesis streams to load data into Amazon Kinesis Analytics. An Amazon Kinesis Analytics application ﬁlters and processes the data and puts it into Amazon Kinesis Firehose. An Amazon Kinesis Firehose delivery stream delivers the analyzed data to various data stores for search, storage, or further analytics. The analyzed data is then put into an Amazon S3 bucket. To persist raw data, AWS Lambda decodes the data and puts it into Amazon Kinesis Firehose which delivers it to Amazon S3.
A real time analytics pipeline was built to enable detection of anomalies in transactions between consumers of internet services.