Chatbot assistant powered by artificial intelligence for troubleshooting issues based on historical resolution data

1 . A system for troubleshooting issues based on historical resolution data comprising:

memory; and

at least one processor, the memory containing software code configured to cause the processor to:

receive source data from a plurality of data sources according to a scheduler that triggers retrieval of the source data using service endpoints and authorization tokens;

extract a plurality of keywords from the source data;

ingest the source data into a staging database, cleanse and normalize the source data, ingest the normalized source data into a production database, and store the extracted plurality of keywords in the production database;

receive a natural-language user issue input from a chatbot;

determine, based on a comparison of the received user issue input to the stored keywords, whether the received user input matches the keywords in the database, wherein the comparison produces a comparison result that is provided to an executable inference artifact deployable for runtime consumption; and

execute the executable inference artifact operating on normalized source data stored in the production database to generate at least one candidate resolution;

when the determination indicates no match, transmit an indication of no resolution and cause the chatbot to request clarification of the user issue input or provide suggested corrected inputs;

in response to the determination, transmit one resolution selected from the generated candidate resolution(s) to the chatbot;

receive feedback associated with a user-selected resolution generated during the chatbot interaction; and

update and store a resolution-ranking data structure based on the received feedback, wherein a subsequent determination of the one resolution includes selecting the one resolution using the stored resolution-ranking data structure.

2 . The system of claim 1 , wherein the data sources comprise one or more of mail data, ticket data, project management data, incident report data.

3 . The system of claim 1 , wherein the data sources are synced with one or more cloud systems or software communication systems.

4 . The system of claim 1 , wherein at least one of the plurality of data sources comprises images;

the processor is further configured to extract image source data with a cognitive vision tool; and

the processor is further configured to store the image source data in the database.

5 . The system of claim 1 , wherein the processor is further configured to convert the source data to JSON.

6 . The system of claim 1 , wherein the resolution ranking data structure is sorted according to at least one of a sorting metric, user feedback, or manual sorting.

7 . The system of claim 1 , wherein the executable inference artifact assigns category identifiers stored in the production database to candidate resolutions.

8 . The system of claim 1 , wherein:

the determining comprises determining that the received user input does not match at least a portion of the stored keywords; and

in response to the determining, transmitting an indication of no resolution.

9 . The system of claim 1 , wherein:

the determining comprises determining that the received user input does match at least a portion of the stored keywords; and

in response to the determining, transmitting the selected resolution.

10 . The system of claim 1 , wherein the processor is further configured to remove undesired source data.

11 . A method for troubleshooting issues based on historical resolution data wherein the processor containing memory is configured to:

receive source data from a plurality of data sources according to a scheduler that triggers retrieval of the source data using service endpoints and authorization tokens;

extract a plurality of keywords from the source data;

ingest the source data into a staging database, cleanse and normalize the source data, ingest the normalized source data into a production database, and store the extracted plurality of keywords in the production database;

receive a natural-language user issue input from a chatbot;

determine, based on a comparison of the received user issue input to the stored keywords, whether the received user input matches the keywords in the database, wherein the comparison produces a comparison result that is provided to an executable inference artifact deployable for runtime consumption; and

execute the executable inference artifact operating on normalized source data stored in the production database to generate at least one candidate resolution;

when the determination indicates no match, transmit an indication of no resolution and cause the chatbot to request clarification of the user issue input or provide suggested corrected inputs;

in response to the determination, transmit one resolution selected from the generated candidate resolution(s) to the chatbot;

receive feedback associated with a user-selected resolution generated during the chatbot interaction; and

update and store a resolution-ranking data structure based on the received feedback, wherein a subsequent determination of the one resolution includes selecting the one resolution using the stored resolution-ranking data structure.

12 . The method of claim 11 , wherein the data sources comprise one or more of mail data, ticket data, project management data, incident report data.

13 . The method of claim 11 , wherein the data sources are synced with one or more cloud systems or software communication systems.

14 . The method of claim 11 , wherein at least one of the plurality of data sources comprises images;

the processor is further configured to extract image source data with a cognitive vision tool; and

the processor is further configured to store the image source data in the database.

15 . The method of claim 11 , wherein the processor is further configured to convert the source data to JSON.

16 . The method of claim 11 , wherein the resolution ranking data structure is sorted according to at least one of a sorting metric, user feedback, or manual sorting.

17 . The method of claim 11 , wherein the executable inference artifact is configured to map at least one of a project name or the keywords to a category.

18 . The method of claim 11 , wherein:

the determining comprises determining that the received user input does not match at least a portion of the stored keywords; and

in response to the determining, transmitting an indication of no resolution.

19 . The method of claim 11 , wherein:

the determining comprises determining that the received user input does match at least a portion of the stored keywords; and

in response to the determining, transmitting the selected resolution.

20 . The method of claim 11 , wherein the processor is further configured to remove undesired source data.