Abstract: A method for aspect-based sentiment analysis includes receiving a collection of textual data, extracting a set of aspects and a set of sentiment words from the textual data, identifying a set of aspect-sentiment word pairs from the extracted aspects and sentiment words, identifying a subset of aspect-sentiment word pairs according to a set of predefined rules, and grouping a plurality of aspects associated with the subset of aspect-sentiment word pairs into one or more clusters. Each of the set of aspect-sentiment word pairs includes an aspect word from the set of aspects and a sentiment word from the set of sentiment words. Each of the subset of aspect-sentiment word pairs is determined to have an aspect-sentiment relationship according to the set of predefined rules.

Abstract: A method for providing document category recommendations may include training a machine learning algorithm; receiving, from a user, a selection of an anchor document; retrieving a user co-viewing sequence; generating, via the trained machine learning algorithm, a sequence embeddings set based on the anchor document and the user co-viewing sequence; comparing the generated embeddings set to respective embeddings for a plurality of candidate sets; determining the candidate set closest to the generated embeddings; and presenting at least one category from the closest candidate set.

Abstract: A method and system for generating and correcting chatbot responses based on reinforcement learning (RL) are disclosed. In some embodiments, the method includes receiving user data associated with a user in a chatbot conversation. The method includes providing a first recommendation to the user. The method includes detecting user feedback to the first recommendation in the chatbot conversation. The method then includes determining whether to assign a positive reward or a negative reward to the user feedback based on sentiment analysis performed on the user feedback. If the negative reward is assigned to the user feedback, the method further includes calculating a negative reward score for the first recommendation; retraining the one or more of RL models using one or more of the negative reward score, the user data, the first recommendation, and the user feedback; and determining a second recommendation using the one or more retrained RL models.

Abstract: A method and system for generating and correcting chatbot responses based on reinforcement learning (RL) are disclosed. In some embodiments, the method includes receiving user data associated with a user in a chatbot conversation. The method includes providing a first recommendation to the user. The method includes detecting user feedback to the first recommendation in the chatbot conversation. The method then includes determining whether to assign a positive reward or a negative reward to the user feedback based on sentiment analysis performed on the user feedback. If the negative reward is assigned to the user feedback, the method further includes calculating a negative reward score for the first recommendation; retraining the one or more of RL models using one or more of the negative reward score, the user data, the first recommendation, and the user feedback; and determining a second recommendation using the one or more retrained RL models.