System and method for electronic text classification
AssigneeWalmart Apollo, LLC
InventorsGyan PRABHAT, Riyanka BHOWAL, Debanjana BANERJEE
Life of this patent
Estimating the term…
Abstract
Systems, method, and computer-readable mediums for automated text classification, and particularly a mechanism for performing binary classification using only a set of positive labeled data as training data and having a large set of unlabeled data, where the algorithm can function without any information regarding the negative class. The disclosed classification systems and methods may use a text classification process which automatically classifies text based on the current positive training data available, but identifies additional words which can be added to the positive training data such that future iterations of the text classification can better identify the positive class of text.
Claims20
A method comprising:
- receiving a plurality of electronic text documents, wherein the plurality of electronic text documents are divided into a labeled portion and an unlabeled portion, the unlabeled portion being larger in quantity than the labeled portion;
- calculating, via a processor, a metric for each respective electronic text document in the labeled portion of the plurality of text documents;
- performing, via the processor, a preliminary classification of each electronic text document in the unlabeled portion of the plurality of electronic text documents based on the metric, to yield preliminary classified data;
- ranking, via the processor, the preliminary classified data in a descending order based on the metric;
- and
- performing, via the processor, a second classification on the preliminary classified data in the descending order.
A system comprising:
- a processor;
- and
- a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
- receiving a plurality of text documents, wherein the plurality of text documents are divided into a labeled portion and an unlabeled portion, the unlabeled portion being larger in quantity than the labeled portion;
- calculating a metric for each respective text document in the labeled portion of the plurality of text documents;
- performing a preliminary classification of each text document in the unlabeled portion of the plurality of text documents based on the metric, to yield preliminary classified data;
- ranking the preliminary classified data in a descending order based on the metric;
- and
- performing a second classification on the preliminary classified data in the descending order.
A non-transitory computer-readable storage medium having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
- receiving a plurality of text documents, wherein the plurality of text documents are divided into a labeled portion and an unlabeled portion, the unlabeled portion being larger in quantity than the labeled portion;
- calculating a metric for each respective text document in the labeled portion of the plurality of text documents;
- performing a preliminary classification of each text document in the unlabeled portion of the plurality of text documents based on the metric, to yield preliminary classified data;
- ranking the preliminary classified data in a descending order based on the metric;
- and
- performing a second classification on the preliminary classified data in the descending order.
Dependencies are inferred from the claim text.