Method and system for aspect-level sentiment classification by graph diffusion transformer

A method comprising:

• receiving, by a computing device, a sentence having an aspect term and context, the aspect term having a classification label;
• converting, by the computing device, the sentence into a dependency tree graph;
• calculating, by the computing device, an attention matrix of the dependency tree graph based on one-hop attention between any two of the plurality of the nodes in the dependency tree graph;
• calculating, by the computing device, graph attention diffusion from multi-hop attention between any two of the plurality of nodes in the dependency tree graph based on the attention matrix;
• obtaining, by the computing device, updated embedding of the dependency tree graph using the graph attention diffusion;
• classifying, by the computing device, the aspect term based on the updated embedding of the dependency tree graph to obtain predicted classification of the aspect term;
• calculating, by the computing device, loss function based on the predicted classification of the aspect term and the classification label of the aspect term;
• and
• adjusting, by the computing device, parameters of models in the computer executable code.

A non-transitory computer readable medium storing computer executable code, wherein the computer executable code, when executed at a processor of an active computing device, is configured to:

• receive a sentence having an aspect term and context, the aspect term having a classification label;
• convert the sentence into a dependency tree graph;
• calculate an attention matrix of the dependency tree graph based on one-hop attention between any two of a plurality of nodes in the dependency tree graph;
• calculate graph attention diffusion from multi-hop attention between any two of the plurality of nodes in the dependency tree graph based on the attention matrix;
• obtain updated embedding of the dependency tree graph using the graph attention diffusion;
• classify the aspect term based on the updated embedding of the dependency tree graph to obtain predicted classification of the aspect term;
• calculate loss function based on the predicted classification of the aspect term and the classification label of the aspect term;
• and
• adjust parameters of models in the computer executable code.