Systems and methods for configurable messaging with feature extraction
Systems and methods for a configurable response-action engine are provided. Actions are generated for a conversation when an insight is received from a natural language processing system. Industry, segment, client specific instructions, third party data, a state for the lead and lead historical patterns are also received. A decision making action model is tuned using this information. An objective for the conversation may be extracted from the state information for the lead. The tuned model is then applied to the insight and objective to output an action. A response message may be generated for the action. The action is directed to cause a state transition of the lead to a preferred state. In another embodiment, systems and methods are presented for feature extraction from one or more messages. In yet other embodiments, systems and methods for message cadence optimization are provided.
1. A computer implemented method for feature extraction comprising:
receiving a plurality of messages;
identifying a set of desired features to be extracted from the messages, wherein each feature in the set is extracted using a transform;
grouping the messages into subsets by conversation, client, industry, conversation state, language and medium, wherein the subsets have common features that are to be extracted;
identifying normalization steps common to each subset, wherein the normalization steps are additive and hierarchical;
normalizing each subset batchwise using the identified normalization steps;
serially applying each transform in a chained manner to the normalized subsets of messages to extract the features
saving after each transform; and
outputting the features to a classification engine for insight extraction.
2. The method of claim 1 , wherein the transforms include text-to-text transforms and text-to-feature transforms.
3. The method of claim 2 , wherein the text-to-text transforms include message cleaning, message normalization, message lemmatize and name-entity-replacement.
4. The method of claim 2 , wherein the text-to-feature transforms include sentence extraction, n-gram production, noun-phrases extraction, and out-of-office features.
5. The method of claim 1 , wherein each transform in the serially applied transforms is completed before the next transform in the serial applied transforms is begun.
6. The method of claim 5 , wherein after each transform in the serial application, the messages are saved with a unique version.
7. The method of claim 6 , further comprising receiving a new feature for extraction from the messages.
8. The method of claim 7 , further comprising applying a new transform associated with the new feature to one of the unique versions to generate an updated version.
9. The method of claim 1 , further comprising annotating the extracted features with facts.
10. The method of claim 9 , wherein the facts are extracted from the messages using natural language processing.