Embedded context extraction using natural language models for dynamic remediation
Method and apparatus for dynamic remediation. A set of records associated with a project is accessed. The set of records is processed using one or more natural language processing techniques to generate textual data comprising a plurality of pairs of sentences corresponding to one or more topics associated with the project. An issue for at least one topic associated with the project is identified based on the textual data, comprising identifying a pair of sentences that comprises a first sentence and a second sentence, calculating a sentence similarity score by comparing the first and second sentences using a similarity metric, and determining that the sentence similarity score satisfies one or more criteria. In response to determining that the one or more criteria are satisfied, a project meeting for the issue is scheduled based at least in part on a criticality of the issue.
1 . A method comprising:
accessing a set of records associated with a project;
processing the set of records, using one or more natural language processing (NLP) techniques, to generate textual data comprising a plurality of pairs of sentences corresponding to one or more topics associated with the project;
identifying an issue for at least one topic associated with the project based on the textual data, comprising:
identifying a pair of sentences from the plurality of pairs of sentences, wherein the pair of sentences comprises a first sentence and a second sentence,
converting the first sentence into a first sentence vector and converting the second sentence into a second sentence vector using a language transformer machine learning (ML) model based on identifying respective part-of-speech (POS) tags of the first and second sentences,
calculating a sentence similarity score, using a similarity metric, by comparing the first and second sentence vectors, and
determining that the sentence similarity score satisfies one or more criteria; and
in response to determining that the one or more criteria are satisfied, scheduling a project meeting for the issue based at least in part on a criticality of the issue.
2 . The method of claim 1 , furthering comprising delineating the textual data into a plurality of dialogues, wherein each of the plurality of dialogues corresponds to a respective topic of a plurality of topics identified for the project.
3 . The method of claim 1 , wherein scheduling the project meeting for the issue comprises:
sending invitation links of the project meeting to a plurality of participants identified based on the textual data, and
generating an ordered agenda comprising the issue identified for the at least one topic and one or more relevant records.
4 . The method of claim 1 , wherein the set of records comprises at least one of: (i) an audio record, or (ii) a video record, or (iii) a textual record.
5 . The method of claim 1 , wherein the similarity metric comprises one of:
(i) an angle similarity metric, or (ii) a distance similarity metric.
6 . The method of claim 1 , wherein converting the first sentence into the first sentence vector, and converting the second sentence into the second sentence vector comprises:
extracting a plurality of features from the first and second sentences,
generating the first sentence vector based on a first set of values using the language transformer ML model, wherein each respective value of the first set of values corresponds to a respective feature of the plurality of features and is identified by processing the first sentence, and
generating the second sentence vector based on a second set of values using the language transformer ML model, wherein each respective value of the second set of values corresponds to a respective feature of the plurality of features and is identified by processing the second sentence.
7 . The method of claim 1 , furthering comprising:
identifying a plurality of issues for the project by processing the textual data, wherein each respective issue of the plurality of issues is assigned a respective criticality, scheduling a projecting meeting for the plurality of issues, further comprising:
sending invitation links of the project meeting to a plurality of participants identified based on the textual data, and
generating an ordered agenda comprises the plurality of issues for the project, wherein the plurality of issues are ordered based on the respective criticality corresponding to each respective issue.
8 . The method of claim 1 , wherein the one or more NLP techniques comprises at least one of: (i) speech-to-text transcription; (ii) keyword extraction and matching; (iii) pattern identification; or (iv) text classification and clustering.
9 . A system, comprising:
one or more memories collectively containing one or more programs; and
one or more processors, wherein the one or more processors are configured to, individually or collectively, perform an operation comprising:
accessing a set of records associated with a project;
processing the set of records, using one or more natural language processing (NLP) techniques, to generate textual data comprising a plurality of pairs of sentences corresponding to one or more topics associated with the project;
identifying an issue for at least one topic associated with the project based on the textual data, comprising:
identifying a pair of sentences from the plurality of pairs of sentences, wherein the pair of sentences comprises a first sentence and a second sentence,
converting the first sentence into a first sentence vector and converting the second sentence into a second sentence vector using a language transformer machine learning (ML) model based on identifying respective part-of-speech (POS) tags of the first and second sentences,
calculating a sentence similarity score, using a similarity metric, by comparing the first and second sentence vectors, and
determining that the sentence similarity score satisfies one or more criteria; and
in response to determining that the one or more criteria are satisfied, scheduling a project meeting for the issue based at least in part on a criticality of the issue.
10 . The system of claim 9 , wherein scheduling the project meeting for the issue comprises:
sending invitation links of the project meeting to a plurality of participants identified based on the textual data, and
generating an ordered agenda comprising the issue identified for the at least one topic and one or more relevant records.
11 . The system of claim 9 , wherein the one or more NLP techniques comprises at least one of: (i) speech-to-text transcription; (ii) keyword extraction and matching; (iii) pattern identification; or (iv) text classification and clustering.
12 . The system of claim 9 , furthering comprising:
identifying a plurality of issues for the project by processing the textual data, wherein each respective issue of the plurality of issues is assigned a respective criticality,
scheduling a projecting meeting for the plurality of issues, further comprising:
sending invitation links of the project meeting to a plurality of participants identified based on the textual data, and
generating an ordered agenda comprises the plurality of issues for the project, wherein the plurality of issues are ordered based on the respective criticality corresponding to each respective issue.
13 . The system of claim 9 , wherein the operation further comprises delineating the textual data into a plurality of dialogues, wherein each of the plurality of dialogues corresponds to a respective topic of a plurality of topics identified for the project.
14 . The system of claim 9 , wherein the similarity metric comprises one of: (i) an angle similarity metric, or (ii) a distance similarity metric.
15 . The system of claim 9 , wherein the set of records comprises at least one of: (i) an audio record, or (ii) a video record, or (iii) a textual record.
16 . The system of claim 9 , wherein converting the first sentence into the first sentence vector, and converting the second sentence into the second sentence vector comprises:
extracting a plurality of features from the first and second sentences,
generating the first sentence vector based on a first set of values using the language transformer ML model, wherein each respective value of the first set of values corresponds to a respective feature of the plurality of features and is identified by processing the first sentence, and
generating the second sentence vector based on a second set of values using the language transformer ML model, wherein each respective value of the second set of values corresponds to a respective feature of the plurality of features and is identified by processing the second sentence.
17 . A computer program product for scheduling project meetings, the computer program product comprising:
a computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by one or more computer processors to:
access a set of records associated with a project;
process the set of records, using one or more natural language processing (NLP) techniques, to generate textual data comprising a plurality of pairs of sentences corresponding to one or more topics associated with the project;
identify an issue for at least one topic associated with the project based on the textual data, comprising:
identifying a pair of sentences from the plurality of pairs of sentences, wherein the pair of sentences comprises a first sentence and a second sentence,
converting the first sentence into a first sentence vector and converting the second sentence into a second sentence vector using a language transformer machine learning (ML) model based on identifying respective part-of-speech (POS) tags of the first and second sentences,
calculating a sentence similarity score, using a similarity metric, by comparing the first and second sentence vectors, and
determining that the sentence similarity score satisfy one or more criteria; and
in response to determining that the one or more criteria are satisfied, schedule a project meeting for the issue based at least in part on a criticality of the issue.
18 . The computer program product of claim 17 , wherein the computer-readable program code executable by the one or more computer processors to schedule the project meeting for the issue further comprising:
sending invitation links of the project meeting to a plurality of participants identified based on the textual data, and
generating an ordered agenda comprising the issue identified for the at least one topic and one or more relevant records.
19 . The computer program product of claim 17 , wherein converting the first sentence into the first sentence vector, and converting the second sentence into the second sentence vector comprises:
extracting a plurality of features from the first and second sentences,
generating the first sentence vector based on a first set of values using the language transformer ML model, wherein each respective value of the first set of values corresponds to a respective feature of the plurality of features and is identified by processing the first sentence, and
generating the second sentence vector based on a second set of values using the language transformer ML model, wherein each respective value of the second set of values corresponds to a respective feature of the plurality of features and is identified by processing the second sentence.
20 . The computer program product of claim 17 , wherein the one or more NLP techniques comprises at least one of: (i) speech-to-text transcription; (ii) keyword extraction and matching; (iii) pattern identification; or (iv) text classification and clustering.