Abstract: The present disclosure relates to a system, a method, and a product for intent discovery. The system includes a processor in communication with a memory storing instructions. When the processor executes the instructions, the instructions are configured to cause the processor to: obtain documents comprising a set of utterances, extract the set of utterances from the documents, generate a set of utterance embeddings based on the set of utterances, clusterize the set of utterance embeddings to obtain a plurality of clusters, obtain a cluster label for each cluster, encode each document based on a number of times each utterance cluster identifier (ID) appears to obtain an encoded document, perform latent Dirichlet allocation (LDA) on the encoded documents to obtain K topics, and each topic corresponding to a list of key clusters with cluster IDs, and for each topic, replace the cluster IDs with the cluster labels.

Abstract: In some implementations, a system may receive non-deterministic finite state automata (NFSA) to represent a set of dialog flows associated with a human-machine interface. The system may generate a deterministic finite state automaton (DFSA) that includes a minimum set of states that represents all dialog flows included in the set of dialog flows represented in the NFSA and does not represent any dialog flows that are not included in the set of dialog flows represented in the NFSA. The system may traverse the DFSA to identify a set of K paths that have a highest total weight based on a weight assigned to each transition in the DFSA. The system may prune the DFSA to remove any states and any transitions that do not belong to the set of K paths. The system may generate an output related to one or more subsets of the set of K paths.