Techniques for negative entity aware augmentation
Novel techniques are described for negative entity-aware augmentation using a two-stage augmentation to improve the stability of the model to entity value changes for intent prediction. In some embodiments, a method comprises accessing a first set of training data for an intent prediction model, the first set of training data comprising utterances and intent labels; applying one or more negative entity-aware data augmentation techniques to the first set of training data, depending on the tuning requirements for hyper-parameters, to result in a second set of training data, where the one or more negative entity-aware data augmentation techniques comprise Keyword Augmentation Technique (“KAT”) plus entity without context technique and KAT plus entity in random context as OOD technique; combining the first set of training data and the second set of training data to generate expanded training data; and training the intent prediction model using the expanded training data.
1 . A method, comprising:
accessing a first set of training data, the first set of training data comprising utterances;
extracting named entities from the first set of training data using a trained Named Entity Recognition (NER) model;
applying one or more negative entity-aware data augmentation techniques to the first set of training data, depending on tuning requirements for hyper-parameters, to result in a second set of training data;
wherein the one or more negative entity-aware data augmentation techniques comprise Keyword Augmentation Technique (“KAT”) plus entity without context technique and KAT plus entity in random context as OOD technique;
combining the first set of training data and the second set of training data to generate expanded training data;
training a machine learning model using the expanded training data to result in a trained machined learning model; and
deploying the trained machine learning model.
2 . The method of claim 1 , further comprising applying Stop Word Augmentation Technique (“SWAT”) by replacing non-stop words in the first set of training data with random stop words while preserving existing stop words to result in the second set of training data.
3 . The method of claim 1 , wherein the KAT plus entity without context technique comprises removing in-domain context information around the named entities within the first set of training data resulting in a modified data as an out-of-domain training data.
4 . The method of claim 3 , wherein vector distance between the out-of-domain training data and any in-domain training data is above a threshold value.
5 . The method of claim 1 , wherein the KAT plus entity in random context as OOD technique comprises replacing context around entity values within the first set of training data with random out-of-domain context.
6 . The method of claim 1 , wherein the one or more negative entity-aware data augmentation techniques train the machine learning model to focus on overall context rather than changes to an entity value of the utterances.
7 . The method of claim 6 , wherein the entity value comprises one or more words representing an individual named entity within a named entity category.
8 . A non-transitory computer-readable medium storing computer-executable instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
accessing a first set of training data, the first set of training data comprising utterances;
extracting named entities from the first set of training data using a trained NER model;
applying one or more negative entity-aware data augmentation techniques to the first set of training data, depending on tuning requirements for hyper-parameters, to result in a second set of training data;
wherein the one or more negative entity-aware data augmentation techniques comprise Keyword Augmentation Technique (“KAT”) plus entity without context technique and KAT plus entity in random context as OOD technique;
combining the first set of training data and the second set of training data to generate expanded training data;
training a machine learning model using the expanded training data to result in a trained machined learning model; and
deploying the trained machine learning model.
9 . The non-transitory computer-readable medium of claim 8 , further comprising applying Stop Word Augmentation Technique (“SWAT”) by replacing non-stop words in the first set of training data with random stop words while preserving existing stop words to result in the second set of training data.
10 . The non-transitory computer-readable medium of claim 8 , wherein the KAT plus entity without context technique comprises removing in-domain context information around the named entities within the first set of training data resulting in a modified data as an out-of-domain training data.
11 . The non-transitory computer-readable medium of claim 10 , wherein vector distance between the out-of-domain training data and any in-domain training data is above a threshold value.
12 . The non-transitory computer-readable medium of claim 8 , wherein the KAT plus entity in random context as OOD technique comprises replacing context around entity values within the first set of training data with random out-of-domain context.
13 . The non-transitory computer-readable medium of claim 8 , wherein the one or more negative entity-aware data augmentation techniques train the machine learning model to focus on overall context rather than changes to an entity value of the utterances.
14 . The non-transitory computer-readable medium of claim 13 , wherein the entity value comprises one or more words representing an individual named entity within a named entity category.
15 . A system, comprising:
one or more processors; and
one or more non-transitory computer readable media storing computer-executable instructions that, when executed by the one or more processors, cause the system to perform:
accessing a first set of training data, the first set of training data comprising utterances;
extracting named entities from the first set of training data using a trained NER model;
applying one or more negative entity-aware data augmentation techniques to the first set of training data, depending on tuning requirements for hyper-parameters, to result in a second set of training data;
wherein the one or more negative entity-aware data augmentation techniques comprise Keyword Augmentation Technique (“KAT”) plus entity without context technique and KAT plus entity in random context as OOD technique;
combining the first set of training data and the second set of training data to generate expanded training data;
training a machine learning model using the expanded training data to result in a trained machine learning model; and
deploying the trained machine learning model.
16 . The system of claim 15 , further comprising applying Stop Word Augmentation Technique (“SWAT”) by replacing non-stop words in the first set of training data with random stop words while preserving existing stop words to result in the second set of training data.
17 . The system of claim 15 , wherein the KAT plus entity without context technique comprises removing in-domain context information around the named entities within the first set of training data resulting in a modified data as an out-of-domain training data.
18 . The system of claim 17 , wherein vector distance between the out-of-domain training data and any in-domain training data is above a threshold value.
19 . The system of claim 15 , wherein the KAT plus entity in random context as OOD technique comprises replacing context around entity values within the first set of training data with random out-of-domain context.
20 . The system of claim 3 , wherein the one or more negative entity-aware data augmentation techniques train the machine learning model to focus on overall context rather than changes to an entity value of the utterances.