Intelligent method to identify complexity of work artifacts

1. An automated method for predicting a complexity of a task associated with a support ticket, the method comprising:

obtaining a plurality of historical support tickets;

generating training data from the plurality of historical support tickets by extracting text characters from text fields and values from value fields of the plurality of historical support tickets, generating features of the plurality of historical support tickets by vectorizing the extracted text characters, and generating labels of the plurality of historical support tickets using the extracted values;

executing a machine learning algorithm to identify patterns in the training data that map the features and the labels to complexity values of the plurality of historical support tickets to train a complexity model to predict a complexity value of a task associated with the support ticket;

receiving a new support ticket that requests assistance with performance of a new task;

extracting second data from the new support ticket;

converting the second data into a set of vectors using a neural network;

determining a coefficient value of the new support ticket based on the set of vectors;

determining an initial complexity value of the new task associated with the new support ticket using the complexity model;

determining a set of similar support tickets from the plurality of historical support tickets that are similar to the new support ticket based on the coefficient value; and

adjusting the initial complexity value of the new task associated with the new support ticket based on the complexity values of similar support tickets in the set of similar support tickets to predict the complexity value of the new task associated with the new support ticket.

2. The method of claim 1 , wherein generating labels of the plurality of historical support tickets using the extracted values comprises:

normalizing each of the extracted values from the value fields of the plurality of historical support tickets so that the extracted values have a maximum value of one;

setting a weight value for each of the extracted values; and

calculating the labels of the plurality of historical support tickets using the weight value for each of the extracted values.

3. The method of claim 2 , wherein calculating the labels of the plurality of historical support tickets using the weight value for each of the extracted values uses formula: L=W 1 *l 1 +Σ i=2 n W i l i , where L is a label, W is the weight value, l is an extracted value/max value, and n is a number of extracted values.

4. The method of claim 1 , further comprising applying a weight to the complexity values of the similar support tickets in the set of similar support tickets.

5. The method of claim 4 , wherein the weight is based on a reporting time of the similar support tickets in the set of similar support tickets.

6. A system configured to determine a complexity of a task, the system comprising memory for storing instructions, and a processor configured to execute the instructions to:

obtain a plurality of historical support tickets;

generate training data from the plurality of historical support tickets by extracting text characters from text fields and values from value fields of the plurality of historical support tickets, generating features of the plurality of historical support tickets by vectorizing the extracted text characters, and generating labels of the plurality of historical support tickets using the extracted values;

execute a machine learning algorithm to identify patterns in the training data that map the features and the labels to complexity values of the plurality of historical support tickets to train a complexity model to predict a complexity value of a task associated with a support ticket;

receive a new support ticket that requests assistance with performance of a new task;

extract second data from the new support ticket;

convert the second data into a set of vectors using a neural network;

determine a coefficient value of the new support ticket based on the set of vectors;

determine an initial complexity value of the new task associated with the new support ticket using the complexity model;

determine a set of similar support tickets from the plurality of historical support tickets that are similar to the new support ticket based on the coefficient value; and

adjust the initial complexity value of the new task associated with the new support ticket based on the complexity values of similar support tickets in the set of similar support tickets to predict the complexity value of the new task associated with the new support ticket.

7. The system of claim 6 , wherein generating the labels of the plurality of historical support tickets using the extracted values comprises:

normalizing each of the extracted values from the value fields of the plurality of historical support tickets so that the extracted values have a maximum value of one;

setting a weight value for each of the extracted values; and

calculating the labels of the plurality of historical support tickets using the weight value for each of the extracted values.

8. The system of claim 7 , wherein calculating the labels of the plurality of historical support tickets using the weight value for each of the extracted values uses formula: L=W 1 *l 1 +Σ i=2 n W i l i , where L is a label, W is the weight value, l is an extracted value/max value, and n is a number of extracted values.

9. The system of claim 6 , wherein the processor is further configured to execute the instructions to apply a weight to the complexity values of the similar support tickets in the set of similar support tickets.

10. The system of claim 9 , wherein the weight is based on a reporting time of the similar support tickets in the set of similar support tickets.

11. A computer program product for determining a complexity of a task, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of a system to cause the system to:

obtain a plurality of historical support tickets;

generate training data from the plurality of historical support tickets by extracting text characters from text fields and values from value fields of the plurality of historical support tickets, generating features of the plurality of historical support tickets by vectorizing the extracted text characters, and generating labels of the plurality of historical support tickets using the extracted values;

execute a machine learning algorithm to identify patterns in the training data that map the features and the labels to complexity values of the plurality of historical support tickets to train a complexity model to predict a complexity value of a task associated with a support ticket;

receive a new support ticket that requests assistance with performance of a new task;

extract second data from the new support ticket;

convert the second data into a set of vectors using a neural network;

determine a coefficient value of the new support ticket based on the set of vectors;

determine an initial complexity value of the new task associated with the new support ticket using the complexity model;

determine a set of similar support tickets from the plurality of historical support tickets that are similar to the new support ticket based on the coefficient value; and

adjust the initial complexity value of the new task associated with the new support ticket based on the complexity values of similar support tickets in the set of similar support tickets to predict the complexity value of the new task associated with the new support ticket.

12. The computer program product of claim 11 , wherein generating the labels of the plurality of historical support tickets using the extracted values comprises:

normalizing each of the extracted values from the value fields of the plurality of historical support tickets so that the extracted values have a maximum value of one;

setting a weight value for each of the extracted values; and

calculating the labels of the plurality of historical support tickets using the weight value for each of the extracted values.

13. The computer program product of claim 11 , wherein the program instructions executable by the processor of the system further cause the system to apply a weight to the complexity values of the similar support tickets in the set of similar support tickets.

14. The computer program product of claim 13 , wherein the weight is based on a reporting time of the similar support tickets in the set of similar support tickets.