System and method for analyzing wireless communication data
View Patent ↗In general, a system and method for analyzing wireless communication records and for determining optimal wireless communication service plans is disclosed. A transceiver is configured to receive billing information associated with a subscriber of a telecommunications service under a current rate plan. A storage unit stores the billing information. A processor processes the subscriber related billing information to produce organized data having a predefined format. The processor then analyzes the processed data in relation to a plurality of rate plans of a plurality of telecommunications service providers, and determines at least one proposed rate plan that would save the subscriber telecommunication costs relative to the current rate plan. A report of at least one proposed rate plan is then produced and provided to the subscriber, which enables selection of a best telecommunication service provider.
1. A method for processing call detail records, the method comprising the steps of:
receiving a plurality of call detail records, the call detail records related to telecommunication service usage by a telecommunication subscriber during a predetermined period;
producing a plurality of calling profile records from the plurality of call details records, each of the plurality of calling profile records representing a profile of the telecommunication service usage by the telecommunication subscriber during a corresponding predetermined period;
determining a trend of the plurality of calling profile records over the corresponding predetermined periods, wherein the trend is determined by analyzing a training set of calling profile records;
calculating an expected usage profile based on the determined trend, the expected usage profile including a usage distribution for at least one of time-of-day, day-of-week, local, and home-market;
calculating an estimated cost for each of a plurality of rate plans based on the expected usage profile; and
selecting one or more rate plans based on the estimated costs;
comparing the selected one or more rate plans to a test set of calling profile records to determine how well the trend that is based on the training set of calling profile records would have predicted the test set.
2. The method of claim 1 , further comprising the steps of:
calculating an estimated cost for each of a plurality of rate plans based on a calling profile record; and
selecting one or more rate plans based on the estimated costs.
3. The method of claim 2 , wherein one of the plurality of rate plans comprises a rate plan currently in service for the telecommunication subscriber.
4. The method of claim 1 , further comprising the step of calculating an average of the calling profile records over the plurality of predetermined periods.
5. The method of claim 1 , further comprising the steps of:
calculating an estimated cost for each of a plurality of rate plans, each estimated cost calculated from an average of the calling profile records over the plurality of predetermined periods; and
determining one or more cost-effective rate plans based on the estimated cost for each rate plan.
6. The method of claim 1 , wherein the predetermined period is equal to one billing period.
7. The method of claim 1 , wherein each of the plurality of call profile records corresponds to a predetermined sub-period, the predetermined sub-period contained within the predetermined period.
8. A method for processing call detail records, the method comprising:
receiving a plurality of call detail records, the call detail records related to telecommunication service usage by a telecommunication subscriber during a predetermined period;
producing a plurality of calling profile records from the plurality of call details records, each of the plurality of calling profile records corresponding to a predetermined period;
determining a trend of the plurality of calling profile records over the corresponding predetermined periods, wherein the trend is determined by analyzing a training set of calling profile records;
calculating an expected usage profile based on the determined trend, the expected usage profile including a usage distribution for at least one of time-of-day, day-of-week, local, and home-market;
calculating an estimated cost for each of a plurality of rate plans based on the expected usage profile;
selecting one or more rate plans based on the estimated costs; and
comparing the selected one or more rate plans to a test set of calling profile records to determine how well the trend that is based on the training set of calling profile records would have predicted the test set.
9. The method of claim 8 , further comprising the steps of:
calculating an estimated cost for each of a plurality of rate plans based on a calling profile record; and
selecting one or more rate plans based on the estimated costs.
10. The method of claim 9 , wherein one of the plurality of rate plans comprises a rate plan currently in service for the telecommunication subscriber.
11. The method of claim 1 , further comprising the step of:
weighting the predetermined periods according to recency during the calculation of the expected usage profile.
12. The method of claim 1 , further comprising the step of:
weighting the predetermined periods, using an exponential envelope function, according to recency during the calculation of the expected usage profile.
13. A method for processing call detail records comprising the steps of:
receiving a plurality of call detail records, the call detail records related to telecommunication service usage by a telecommunication subscriber during a predetermined period; and;
producing a plurality of calling profile records from the plurality of call details records, each of the plurality of calling profile records representing a profile of the telecommunications service usage by the telecommunication subscriber during a corresponding predetermined period;
determining a trend of the plurality of calling profile records over the corresponding predetermined periods, wherein the trend is determined by analyzing a training set of calling profile records;
calculating an expected usage profile based on the determined trend;
calculating an estimated cost for each of a plurality of available rate plans based on the expected usage profile;
determining whether insufficient data exists for selecting an optimal rate plan;
responsive to the determination that insufficient data exists for selecting an optimal rate plan, notifying the telecommunication subscriber that insufficient data exists for selecting an optimal rate plan;
determining, based on the estimated cost for each of the available rate plans, whether a current rate plan is optimal;
responsive to the determination that the current rate plan is not optimal, determining based on the estimated cost for each of the available rate plans, whether one of the available rate plans represents a cost reduction exceeding a predetermined threshold as compared to the current rate plan; and
responsive to the determination that one of the available rate plans represents a cost reduction, notifying the telecommunication subscriber about the one of the available rate plans representing the cost reduction; and
comparing the one or more rate plans to a test set of calling profile records to determine how well the trend that is based on the training set of calling profile records would have predicted the test set.
14. The method of claim 13 , further comprising the steps of:
determining the number of corresponding predetermined periods; and
determining that insufficient data exists for selecting an optimal rate plan when the number of corresponding predetermined periods is one.
15. The method of claim 14 , further comprising the steps of:
determining a most recent period for the calling profile records; and
calculating a standard deviation in the most recent period; and
determining that insufficient data exists for selecting an optimal rate plan when the standard deviation exceeds one sigma.
16. The method of claim 14 , wherein the expected usage profile includes a usage distribution for at least one of time-of-day, day-of-week, local, and home-market.