Enhanced survey information synthesis
Enhanced survey information synthesis can include performing a respondent assessment of a survey respondent based on respondent data obtained electronically from one or more electronic data sources. Survey responses provided by the survey respondent can be adjusted, the adjusting based on the respondent assessment. A revised survey can be generated, the revised survey comprising the survey responses adjusted based on the respondent assessment.
1. A computer program product, comprising
a computer readable hardware storage device having program instructions embodied therewith,
the program instructions, which when executed by a computer hardware system including a respondent assessor, a response adjuster assessor, a source assessor, and a revised results generator, cause the computer hardware system to perform:
receiving, from a survey source and by the source assessor, electronic survey data corresponding to a survey including survey responses from a plurality of survey respondents;
retrieving, for each of the plurality of survey respondents and by the source assessor, respondent data from an electronic data source separate from the computer hardware system;
performing a respondent assessment, using a deep learning neural network within the respondent assessor, of the survey respondent based on the respondent data;
adjusting, using a classification generated by the deep learning neural network for each of the plurality of survey respondents and by the response adjuster, the survey responses to generate adjusted survey responses; and
generating, by the revised results generator, revised electronic survey data corresponding to a revised survey including the adjusted survey responses, wherein
the survey source is an electronic device, wherein
the deep learning neural network is configured to:
generate an output based upon an input vector fed through a plurality of layers of the deep learning neural network, and
adjust weights of the plurality of layers by applying a backpropagation algorithm to the output.
2. The computer program product of claim 1 , wherein
the deep learning neural network is a classification model constructed using machine learning.
3. The computer program product of claim 2 , wherein
the classification model is configured to classify each of the plurality of survey respondents based on one or more respondent-specific factors.
4. The computer program product of claim 1 , wherein
the deep learning neural network is configured to receive, as input, a feature vector corresponding to a specific survey respondent of the plurality of survey respondents, and
each element of the feature vector corresponds to an attribute of the specific survey respondent.
5. The computer program product of claim 4 , wherein
at least one element of the feature vector is based upon an electronic message posted by the specific survey respondent.
6. The computer program product of claim 1 , wherein
an output of the deep learning neural network is a vector having elements indicating a probability that a specific survey respondent of the plurality of survey respondents belong to a specific classification.
7. The computer program product of claim 1 , wherein
the adjusting includes weighting a particular survey response for a particular survey respondent based upon an analysis of the particular survey respondent using a tone analyzer.
8. A computer hardware system, comprising:
a hardware processor including a respondent assessor, a response adjuster assessor, a source assessor, and a revised results generator and configured to initiate the following operations:
receiving, from a survey source and by the source assessor, electronic survey data corresponding to a survey including survey responses from a plurality of survey respondents;
retrieving, for each of the plurality of survey respondents and by the source assessor, respondent data from an electronic data source separate from the computer hardware system;
performing a respondent assessment, using a deep learning neural network within the respondent assessor, of the survey respondent based on the respondent data;
adjusting, using a classification generated by the deep learning neural network for each of the plurality of survey respondents and by the response adjuster, the survey responses to generate adjusted survey responses; and
generating, by the revised results generator, revised electronic survey data corresponding to a revised survey including the adjusted survey responses, wherein the survey source is an electronic device, wherein
the deep learning neural network is configured to:
generate an output based upon an input vector fed through a plurality of layers of the deep learning neural network, and
adjust weights of the plurality of layers by applying a backpropagation algorithm to the output.
9. The system of claim 8 , wherein
the deep learning neural network is a classification model constructed using machine learning.
10. The system of claim 9 , wherein
the classification model is configured to classify each of the plurality of survey respondents based on one or more respondent-specific factors.
11. The system of claim 8 , wherein
the deep learning neural network is configured to receive, as input, a feature vector corresponding to a specific survey respondent of the plurality of survey respondents, and
each element of the feature vector corresponds to an attribute of the specific survey respondent.
12. The system of claim 11 , wherein
at least one element of the feature vector is based upon an electronic message posted by the specific survey respondent.
13. The system of claim 8 , wherein
an output of the deep learning neural network is a vector having elements indicating a probability that a specific survey respondent of the plurality of survey respondents belong to a specific classification.
14. The system of claim 8 , wherein
the adjusting includes weighting a particular survey response for a particular survey respondent based upon an analysis of the particular survey respondent using a tone analyzer.