Socio-mindfulness in multi-party discussions
Techniques are described with respect to a system, method, and computer program product for predicting conversation for multi-party discussions. An associated method includes analyzing a plurality of linguistic inputs; generating an interest graph based on the analysis; extracting a plurality of temporal data from the interest graph; and generating a Post-Salutations Alignment Model (PSAM) based on clustering of the plurality of temporal data.
1 . A computer-implemented method for predicting conversation for multi-party discussions, the method comprising:
analyzing, by a computing device, a plurality of linguistic inputs;
generating, by the computing device, an interest graph based on the analysis;
extracting, by the computing device, a plurality of temporal data from the interest graph; and
generating, by the computing device, a Post-Salutations Alignment Model (PSAM) based on clustering of the plurality of temporal data via Gaussian Mixture Models (GMM) for Expectation-Maximization (EM) clustering.
2 . The computer-implemented method of claim 1 further comprising:
utilizing, by the computing device, the PSAM to generate at least one conversational starter associated with a multi-party discussion.
3 . The computer-implemented method of claim 1 , wherein clustering of the plurality of temporal data comprises:
utilizing, by the computing device, the GMM for Expectation-Maximization (EM) clustering to discover a plurality of trends and noise within the plurality of linguistic inputs.
4 . The computer-implemented method of claim 1 , wherein generating the PSAM comprises:
performing, by the computing device, topic modeling by using Latent Dirichlet Allocation (LDA).
5 . The computer-implemented method of claim 4 , wherein the topic modeling is based on a plurality of results of the EM clustering.
6 . The computer-implemented method of claim 1 , wherein analyzing the plurality of linguistic inputs comprises:
utilizing, by the computing device, Convolutional Neural Networks (CNN) for one or more of text classification and user intent classification associated with the plurality of linguistic inputs.
7 . The computer-implemented method of claim 6 , wherein text classification and user intent classification comprise:
detecting, by the computing device, a plurality of key phrases associated with the plurality of linguistic inputs.
8 . A computer program product for predicting conversation for multi-party discussions, the computer program product comprising one or more computer readable storage media and program instructions collectively stored on the one or more computer readable storage media, the stored program instructions comprising:
program instructions to analyze a plurality of linguistic inputs;
program instructions to generate an interest graph based on the analysis;
program instructions to extract a plurality of temporal data from the interest graph; and
program instructions to generate a Post-Salutations Alignment Model (PSAM) based on clustering of the plurality of temporal data via Gaussian Mixture Models (GMM) for Expectation-Maximization (EM) clustering.
9 . The computer program product of claim 8 , the stored program instructions further comprising:
program instructions to utilize the PSAM to generate at least one conversational starter associated with a multi-party discussion.
10 . The computer program product of claim 8 , wherein program instructions to cluster the plurality of temporal data comprise:
program instructions to utilize the Gaussian Mixture Models (GMM) for Expectation-Maximization (EM) clustering to discover a plurality of trends and noise within the plurality of linguistic inputs.
11 . The computer program product of claim 8 , wherein program instructions to generate the PSAM comprise:
program instructions to perform topic modeling by using Latent Dirichlet Allocation (LDA).
12 . The computer program product of claim 11 , wherein the topic modeling is based on a plurality of results of the Expectation-Maximization (EM) clustering.
13 . The computer program product of claim 8 , wherein program instructions to analyze the plurality of linguistic inputs comprise:
program instructions to utilize Convolutional Neural Networks (CNN) for one or more of text classification and user intent classification associated with the plurality of linguistic inputs.
14 . The computer program product of claim 13 , wherein text classification and user intent classification comprise:
program instructions to detect a plurality of key phrases associated with the plurality of linguistic inputs.
15 . A computer system for predicting conversation for multi-party discussions, the computer system comprising:
one or more processors;
one or more computer-readable memories;
program instructions stored on at least one of the one or more computer-readable memories for execution by at least one of the one or more processors, the program instructions comprising:
program instructions to analyze a plurality of linguistic inputs;
program instructions to generate an interest graph based on the analysis;
program instructions to extract a plurality of temporal data from the interest graph; and
program instructions to generate a Post-Salutations Alignment Model (PSAM) based on clustering of the plurality of temporal data via Gaussian Mixture Models (GMM) for Expectation-Maximization (EM) clustering.
16 . The computer system of claim 15 , the stored program instructions further comprising:
program instructions to utilize the PSAM to generate at least one conversational starter associated with a multi-party discussion.
17 . The computer system of claim 15 , wherein program instructions to cluster the plurality of temporal data comprise:
program instructions to utilize the Gaussian Mixture Models (GMM) for Expectation-Maximization (EM) clustering to discover a plurality of trends and noise within the plurality of linguistic inputs.
18 . The computer system of claim 15 , wherein program instructions to generate the PSAM comprise:
program instructions to perform topic modeling by using Latent Dirichlet Allocation (LDA).
19 . The computer system of claim 18 , wherein the topic modeling is based on a plurality of results of the Expectation-Maximization (EM) clustering.
20 . The computer system of claim 18 , wherein program instructions to analyze the plurality of linguistic inputs comprise:
program instructions to utilize Convolutional Neural Networks (CNN) for one or more of text classification and user intent classification associated with the plurality of linguistic inputs.