| CPC G06N 3/08 (2013.01) [G06N 3/04 (2013.01); G06N 5/04 (2013.01); G06N 7/01 (2023.01); G06N 20/00 (2019.01)] | 17 Claims |
|
1. A system to automate a resolution process associated with multidimensional data items using predictive analysis, comprising:
a processor; and
a model generator coupled to the processor, the model generator to:
receive, by a communication interface, a resolved data item relating to a service issue via a network, the resolved data item including a set of attributes corresponding to a plurality of predefined data dimensions, wherein the set of attributes relate to a historical resolution for the service issue, a category related to the service issue, a description of the service issue, an identity indicator, and a creation date with a first timestamp related to the resolved data item;
preprocess, by the processor, the resolved data item for generating a preprocessed data, wherein the resolved data item is preprocessed by:
obtaining one of a log-normal and a near normal distribution of the set of attributes by adjusting, each of the set of attributes to a skewed data based on long transformation techniques;
forecasting future states of the set of attributes by evaluating the set of attributes against respective predefined performance metrics using one of moving averages and an autoregressive integrated moving average (ARIMA) technique;
filtering numerical attributes from the set of attributes;
segregating the set of attributes into one of numerical attributes and categorical attributes based on the predefined performance metrics;
obtaining numerical representations of the categorical attributes for statistical analysis based on one of dummy encoding technique and one-hot encoding technique;
normalizing a natural language text for the set of attributes based on one of a sentence tokenization, a word tokenization, a lemmatization, a stemming, a stop word removal, a spell-check, a special character removal, and part-of-speech (POS) tagging techniques; and
normalizing the set of attributes using data sufficiency and accuracy threshold techniques to generate the preprocessed data;
transform, by the processor, the preprocessed data into a metadata corresponding to the plurality of predefined data dimensions based on the predefined performance metrics;
adjust, by the processor, a population of attributes in the set of attributes based on a plurality of data models including a statistical data model and a deep learning data model operating independent of each other,
wherein to adjust the population of the attributes, the processor is configured to:
generate a predictive feature by operating on the set of attributes by the statistical data model,
execute, by the deep learning data model, a grid search to stack multiple layers of the preprocessed data requiring focus on specific categories relative to other categories of attributes of the set of attributes as per the predefined performance metrics, wherein the deep learning data model includes a partial autoencoder that compresses features of the preprocessed data by focusing on the specific categories based on the grid search, and
generate a predictive label based on the predefined performance metrics related to the plurality of predefined data dimensions based on the compressed features, and
wherein the predictive feature and the predictive label collectively define training data;
train, by the processor, a classification model based on the training data, to provide a trained data model predicting a potential issue related to an unresolved data item, wherein the trained data model provides a trigger based on the potential issue being related to the predefined performance metrics used for obtaining the training data; and
an action performer coupled to the processor, in response to the trigger, the action performer to:
classify, by the processor, the unresolved data item into a priority group related to the predefined performance metrics, wherein the priority group corresponds to one of an age, an escalation, and a sentiment, wherein the performance metrics are related to internal and external control objectives of a process, wherein the internal control objectives relate to one of a reliability of a process, a timely update on achievement of operational and strategic goals, and compliance with laws and regulations; and
perform, by the processor, a predefined action based on the priority group associated with the unresolved data item, the predefined action comprising at least one of:
configuring a data model for predicting a resolution for the potential issue;
manipulating a position of the unresolved data item in a predefined queue;
assigning a rank to the unresolved data item based on the priority group;
communicating at least one of the potential issue and the unresolved data item to a predefined device; and
initiating the resolution associated with the priority group and the potential issue.
|