Systems and methods for accessing distributed service systems to surface insights for a user

1 . A method for performing data access and data disambiguation in a distributed processing environment, comprising:

accessing, by a computer processing system, service system data generated by services of a distributed services system for a plurality of users of the distributed services system, the service system data comprising one or more values associated with one or more corresponding common entities;

parsing, by the computer processing system, the service system data to extract data descriptors and associated values from within the service system data that are associated with a set of metrics;

translating, by the computer processing system, two or more different data descriptors associated with a same metric to a common entity, wherein associated values of the two or more different data descriptors are further associated with the common entity;

storing, by the computer processing system, the common entity and values associated with the common entity in a data store of disambiguated service system data;

accessing the data store of the disambiguated service system data to obtain sets of values associated with common entities; and

iteratively inputting the sets of values for the common entities into a machine learning model (MLM) during a training phase of the MLM, the training phase to generate a trained MLM that detects relationships between the common entities and the set of metrics.

2 . The method of claim 1 , wherein the relationships between the common entities and the set of metrics comprises a cohort level relationship where values of the set of metrics used by the trained MLM are for anomaly detection of metrics based on service system data generated by users of the distributed services system having a similar size, similar location, similar organization type, or a combination thereof.

3 . The method of claim 1 , wherein the relationships between the common entities and the set of metrics comprises a trend based variation in values of one or more of the metrics in the set of metrics over a period of time.

4 . The method of claim 1 , further comprising:

accessing service system data generated for a user by services of the distributed service system, the service system data generated for the user comprising one or more values associated with one or more corresponding common entities in the data store of disambiguated service system data;

inputting the one or more values associated with the one or more corresponding common entities into the trained MLM;

detecting, by the trained MLM, an anomaly in a metric in the set of metrics; and

transmitting, by the computer processing system to a second computer processing system associated with the user, a report generated by the computer processing system indicative of the detection of the anomaly in the metric.

5 . The method of claim 4 , wherein the anomaly in the metric comprises an anomaly in a trend of the metric.

6 . The method of claim 4 , wherein the report comprises a dashboard user interface rendered on a display screen of the second computer processing system.

7 . The method of claim 4 , the method further comprising:

identifying, by the computer processing system, at least one entity and an associated value of the at least one entity determined by the trained MLM that cause the anomaly in the metric; and

adding data indicative of the at least one entity and the associated value of the at least one entity to the report prior to transmission to the second computer processing systems.

8 . The method of claim 7 , further comprising:

determining, by the computer processing system, that the at least one entity and the associated value are used to configure a service of the distributed services system; and

adjusting, by the computer processing system, the associated value of the at least one entity at the service of the distributed service system by changing the associated value to a new value that does not cause the anomaly in the metric.

9 . A non-transitory computer readable storage medium having instructions stored thereon, which when executed by a computer processing system, causes the computer processing system to perform operations for performing data access and data disambiguation in a distributed processing environment, the operations comprising:

accessing, by the computer processing system, service system data generated by services of a distributed services system for a plurality of users of the distributed services system, the service system data comprising one or more values associated with one or more corresponding common entities;

parsing, by the computer processing system, the service system data to extract data descriptors and associated values from within the service system data that are associated with a set of metrics;

translating, by the computer processing system, two or more different data descriptors associated with a same metric to a common entity, wherein associated values of the two or more different data descriptors are further associated with the common entity;

storing, by the computer processing system, the common entity and values associated with the common entity in a data store of disambiguated service system data;

accessing the data store of disambiguated service system data to obtain sets of values associated with common entities; and

iteratively inputting the sets of values for the common entities into a machine learning model (MLM) during a training phase of the MLM, the training phase to generate a trained MLM that detects relationships between the common entities and the set of metrics.

10 . The non-transitory computer readable storage medium of claim 9 , wherein the relationships between the common entities and the set of metrics comprises a cohort level relationship where values of the set of metrics used by the trained MLM are for anomaly detection of metrics based on service system data generated by users of the distributed services system having a similar size, similar location, similar organization type, or a combination thereof.

11 . The non-transitory computer readable storage medium of claim 9 , wherein the relationships between the common entities and the set of metrics comprises a trend based variation in values of one or more of the metrics in the set of metrics over a period of time.

12 . The non-transitory computer readable storage medium of claim 9 , the operations further comprising:

accessing service system data generated for a user by services of the distributed service system, the service system data generated for the user comprising one or more values associated with one or more corresponding common entities in the data store of disambiguated service system data;

inputting the one or more values associated with the one or more corresponding common entities into the trained MLM;

detecting, by the trained MLM, an anomaly in a metric in the set of metrics; and

transmitting, by the computer processing system to a second computer processing system associated with the user, a report generated by the computer processing system indicative of the detection of the anomaly in the metric.

13 . The non-transitory computer readable storage medium of claim 12 , the operations further comprising:

identifying, by the computer processing system, at least one entity and an associated value of the at least one entity determined by the trained MLM that cause the anomaly in the metric; and

adding data indicative of the at least one entity and the associated value of the at least one entity to the report prior to transmission to the second computer processing systems.

14 . A system for performing data access and data disambiguation in a distributed processing environment, the system comprising:

a memory; and

at least one processor coupled with the memory configured to:

access service system data generated by services of a distributed services system for a plurality of users of the distributed services system, the service system data comprising one or more values associated with one or more corresponding common entities,

parse the service system data to extract data descriptors and associated values from within the service system data that are associated with a set of metrics,

translate two or more different data descriptors associated with a same metric to a common entity, wherein associated values of the two or more different data descriptors are further associated with the common entity, and

store the common entity and values associated with the common entity in a data store of disambiguated service system data;

access the data store of disambiguated service system data to obtain sets of values associated with common entities; and

iteratively input the sets of values for the common entities into a machine learning model (MLM) during a training phase of the MLM, the training phase to generate a trained MLM that detects relationships between the common entities and the set of metrics.

15 . The system of claim 14 , wherein the relationships between the common entities and the set of metrics comprises a cohort level relationship where values of the set of metrics used by the trained MLM are for anomaly detection of metrics based on service system data generated by users of the distributed services system having a similar size, similar location, similar organization type, or a combination thereof.

16 . The system of claim 14 , wherein the relationships between the common entities and the set of metrics comprises a trend based variation in values of one or more of the metrics in the set of metrics over a period of time.

17 . The system of claim 14 , wherein the at least one processor is further configured to:

access service system data generated for a user by services of the distributed service system, the service system data generated for the user comprising one or more values associated with one or more corresponding common entities in the data store of disambiguated service system data;

input the one or more values associated with the one or more corresponding common entities into the trained MLM;

detect, by the trained MLM, an anomaly in a metric in the set of metrics; and

transmit, to a second computer processing system associated with the user, a report generated by the computer processing system indicative of the detection of the anomaly in the metric.