Guided problem resolution using machine learning
Methods and systems are disclosed that include receiving problem information from a user interface at a resolution identification system, receiving product information at the resolution identification system, and performing machine learning analysis of the problem information and the product information. The machine learning analysis produces one or more model outputs, and is performed by a machine learning system of the resolution identification system, using one or more machine learning models. Each of the one or more machine learning models produces a corresponding one of the one or more model outputs. Such a method can further include generating resolution information by performing an action identification operation using the one or more model outputs, and outputting the resolution information from the resolution identification system. The resolution information is output to the user interface.
1. A method comprising:
receiving problem information from a user interface, wherein
the problem information is received at a resolution identification system, and
the problem information describes one or more characteristics of a problem;
receiving product information at the resolution identification system, wherein
the product information describes one or more characteristics of a product;
training a machine learning system using a logistic regression function;
iteratively performing preprocessing of the problem information until the preprocessed problem information is determined to be in an acceptable level of accuracy as input to the machine learning system, wherein the preprocessing comprises term frequency-inverse document frequency analysis;
performing machine learning analysis of the problem information and the product information, wherein
the machine learning analysis produces one or more model outputs,
the machine learning analysis is performed by the machine learning system of the resolution identification system,
the machine learning analysis is performed using one or more machine learning models, wherein at least one of the one or more machine learning models is trained as a logistic regression model, and
each of the one or more machine learning models produces a corresponding one of the one or more model outputs, wherein at least one of the one or more machine learning models classifies the problem information as a problem type, wherein the one or more machine learning models comprises a guided path model, a soft model and a hard model and wherein the one or more model outputs comprises a guided path solution, a soft solution and a hard solution, wherein the one or more model outputs is associated with a respective guided path solution display, software solution display and hard solution display;
updating business rule information based on the problem type;
generating resolution information by performing an action identification operation using the one or more model outputs; and
outputting the resolution information from the resolution identification system, wherein
the resolution information is output to the user interface, wherein the user interface comprises an individual model predictions display comprising the guided path solution display rendering the guided path solution and a guided path solution confidence level indicator, the software solution display rendering the soft solution and a software solution confidence level indicator, and the hard solution display rendering the hard solution and a hard solution confidence level indicator, and wherein the user interface comprises at least one user feedback field accepting input associated with the resolution information that is outputted by the individual model predictions display.
2. The method of claim 1 , wherein the product information comprises tag information for a product.
3. The method of claim 2 , wherein the problem information comprises at least one of error information regarding an error experienced in operation of the product, or symptom information regarding a symptom exhibited by the product in the operation of the product.
4. The method of claim 1 , further comprising:
retrieving one or more system attributes for a product identified by the product information, and
retrieving a support history for the product.
5. The method of claim 1 , further comprising:
performing an outcome analysis, wherein the outcome analysis is based, at least in part, on the resolution information, the machine learning analysis is performed by one or more machine learning systems of the resolution identification system, and a result of the outcome analysis is fed back to the one or more machine learning systems.
6. The method of claim 1 , further comprising:
performing data processing on the problem information, wherein the data processing comprises generating prepared data by performing a data cleansing operation on the problem information, and performing natural language processing on the prepared data.
7. The method of claim 6 , wherein the performing data processing further comprises:
identifying one or more historically successful resolutions, wherein
the one or more historically successful resolutions are identified by analyzing existing historical data; and
creating a sub-intelligence engine, wherein the sub-intelligence engine is created using a portion of the problem information.
8. The method of claim 1 , wherein the one or more machine learning models comprise at least one of:
a guided path model,
a soft model,
a hard model, or
a cluster model.
9. The method of claim 1 , further comprising:
generating a machine learning input to the one or more machine learning models, wherein the problem information comprises error information, and symptom information, and the machine learning input is generated by generating error code information by performing error code interpretation on the error information, and performing keyword extraction on the error code information and the symptom information.
10. The method of claim 9 , further comprising:
performing an outcome analysis, wherein the machine learning input is one of a plurality of machine learning inputs, the plurality of machine learning inputs comprise existing problem information, and the business rule information, the outcome analysis is based, at least in part, on the resolution information, the machine learning analysis is performed by one or more machine learning systems of the resolution identification system, and results of the outcome analysis are fed back to the existing problem information and the business rule information.
11. A non-transitory computer-readable storage medium comprising program instructions, which, when executed by one or more processors of a computing system, perform a method comprising:
receiving problem information from a user interface, wherein
the problem information is received at a resolution identification system, and
the problem information describes one or more characteristics of a problem;
receiving product information at the resolution identification system, wherein
the product information describes one or more characteristics of a product;
training a machine learning system using a logistic regression function;
iteratively performing preprocessing of the problem information until the preprocessed problem information is determined to be in an acceptable level of accuracy as input to the machine learning system, wherein the preprocessing comprises term frequency-inverse document frequency analysis;
performing machine learning analysis of the problem information and the product information, wherein
the machine learning analysis produces one or more model outputs,
the machine learning analysis is performed by the machine learning system of the resolution identification system,
the machine learning analysis is performed using one or more machine learning models, wherein at least one of the one or more machine learning models is trained as a logistic regression model, and
each of the one or more machine learning models produces a corresponding one of the one or more model outputs, wherein at least one of the one or more machine learning models classifies the problem information as a problem type, wherein the one or more machine learning models comprises a guided path model, a soft model and a hard model and wherein the one or more model outputs comprises a guided path solution, a soft solution and a hard solution, wherein the one or more model outputs is associated with a respective guided path solution display, software solution display and hard solution display;
updating business rule information based on the problem type;
generating resolution information by performing an action identification operation using the one or more model outputs; and
outputting the resolution information from the resolution identification system, wherein
the resolution information is output to the user interface, wherein the user interface comprises an individual model predictions display comprising the guided path solution display rendering the guided path solution and a guided path solution confidence level indicator, the software solution display rendering the soft solution and a software solution confidence level indicator, and the hard solution display rendering the hard solution and a hard solution confidence level indicator, and wherein the user interface comprises at least one user feedback field accepting input associated with the resolution information that is outputted by the individual model predictions display.
12. The non-transitory computer-readable storage medium of claim 11 , wherein the method further comprises:
retrieving one or more system attributes for a product identified by the product information, and
retrieving a support history for the product, wherein
the product information comprises tag information for a product, and
the problem information comprises at least one of
error information regarding an error experienced in operation of the product, or
symptom information regarding a symptom exhibited by the product in the operation of the product.
13. The non-transitory computer-readable storage medium of claim 11 , wherein the method further comprises:
performing an outcome analysis, wherein
the outcome analysis is based, at least in part, on the resolution information,
the machine learning analysis is performed by one or more machine learning systems of the resolution identification system, and
a result of the outcome analysis is fed back to the one or more machine learning systems.
14. The non-transitory computer-readable storage medium of claim 11 , wherein the method further comprises:
performing data processing on the problem information, wherein
the data processing comprises
generating prepared data by performing a data cleansing operation on the problem information, and
performing natural language processing on the prepared data;
identifying one or more historically successful resolutions, wherein
the one or more historically successful resolutions are identified by analyzing existing historical data; and
creating a sub-intelligence engine, wherein
the sub-intelligence engine is created using a portion of the problem information.
15. The non-transitory computer-readable storage medium of claim 11 , wherein the one or more machine learning models comprise at least one of
a guided path model,
a soft model,
a hard model, or
a cluster model.
16. A system comprising:
one or more processors; and
a computer-readable storage medium coupled to the one or more processors, comprising program instructions, which, when executed by the one or more processors, perform a method comprising:
receiving problem information from a user interface, wherein
the problem information is received at a resolution identification system, and
the problem information describes one or more characteristics of a problem;
receiving product information at the resolution identification system, wherein
the product information describes one or more characteristics of a product;
training a machine learning system using a logistic regression function;
iteratively performing preprocessing of the problem information until the preprocessed problem information is determined to be in an acceptable level of accuracy as input to the machine learning system, wherein the preprocessing comprises term frequency-inverse document frequency analysis;
performing machine learning analysis of the problem information and the product information, wherein
the machine learning analysis produces one or more model outputs,
the machine learning analysis is performed by the machine learning system of the resolution identification system,
the machine learning analysis is performed using one or more machine learning models, wherein at least one of the one or more machine learning models is trained as a logistic regression model, and
each of the one or more machine learning models produces a corresponding one of the one or more model outputs, wherein at least one of the one or more machine learning models classifies the problem information as a problem type, wherein the one or more machine learning models comprises a guided path model, a soft model and a hard model and wherein the one or more model outputs comprises a guided path solution, a soft solution and a hard solution, wherein the one or more model outputs is associated with a respective guided path solution display, software solution display and hard solution display;
updating business rule information based on the problem type;
generating resolution information by performing an action identification operation using the one or more model outputs; and
outputting the resolution information from the resolution identification system, wherein
the resolution information is output to the user interface, wherein the user interface comprises an individual model predictions display comprising the guided path solution display rendering the guided path solution and a guided path solution confidence level indicator, the software solution display rendering the soft solution and a software solution confidence level indicator, and the hard solution display rendering the hard solution and a hard solution confidence level indicator, and wherein the user interface comprises at least one user feedback field accepting input associated with the resolution information that is outputted by the individual model predictions display.
17. The system of claim 16 , wherein the method further comprises:
retrieving one or more system attributes for a product identified by the product information, and
retrieving a support history for the product, wherein
the product information comprises tag information for a product, and
the problem information comprises at least one of
error information regarding an error experienced in operation of the product, or
symptom information regarding a symptom exhibited by the product in the operation of the product.
18. The system of claim 16 , wherein the method further comprises:
performing an outcome analysis, wherein
the outcome analysis is based, at least in part, on the resolution information,
the machine learning analysis is performed by one or more machine learning systems of the resolution identification system, and
a result of the outcome analysis is fed back to the one or more machine learning systems.
19. The system of claim 16 , wherein the method further comprises:
performing data processing on the problem information, wherein
the data processing comprises
generating prepared data by performing a data cleansing operation on the problem information, and
performing natural language processing on the prepared data;
identifying one or more historically successful resolutions, wherein
the one or more historically successful resolutions are identified by analyzing existing historical data; and
creating a sub-intelligence engine, wherein
the sub-intelligence engine is created using a portion of the problem information.
20. The system of claim 16 , wherein
the one or more machine learning models comprise at least one of
a guided path model,
a soft model,
a hard model, or
a cluster model.