The present invention relates to the technical field of computers, and provides polysemant meaning learning and search result display. The polysemant meaning learning comprises: extracting a plurality of first target words and at least one adjacent word combination of each first target word from a text set to be learned; respectively encoding each first target word and each adjacent word combination according to a word bank of the text set to be learned; training a capsule network model by using the encoding of each first target word as an input vector and the encoding of each adjacent word combination corresponding to each first target word as an output vector; when recognition is performed on a second target word to be recognized, inputting the second target word in the capsule network model, and determining a plurality of obtained intermediate vectors as feature vectors of the second target word; and clustering the feature vectors with a cosine similarity greater than a similarity threshold to generate a representative word of each category, and determining one or more meanings of each second target word according to one or more categories of representative words to which the feature vectors of the second target word belong.
展开▼
