Systems, computer implementation methods, and computer programs (conversational system content related to external events)

Abstract

To provide a system, computer-implemented method and computer program product for conversational system content related to external events.SOLUTION: The method comprises performing the steps of: receiving chatbot logs; clustering or classifying the chatbot logs into clusters of potential intents; detecting trending topics by analyzing at least social media content; computing semantic similarities between the potential intents and the trending topics; based on the computed semantic similarities, correlating at least one of the potential intents to at least one of the trending topics; and triggering training of the chatbot using the chatbot logs associated with the correlated potential intent.SELECTED DRAWING: Figure 7

Claims

1. Hardware processor and Includes a memory device coupled to the hardware processor, The aforementioned hardware processor comprises at least, Receive the chatbot log, which includes at least the questions received by the chatbot that is conversing with the user. Cluster chatbot logs into clusters of potential intents. At the very least, by analyzing social media content, we can detect trending topics. The semantic similarity between the latent intent and the trend topic is calculated, Based on the calculated semantic similarity, at least one of the potential intents is correlated with at least one of the trend topics. A system configured to initiate training of the chatbot using the chatbot logs related to the correlated potential intents.

2. The system according to claim 1, wherein the hardware processor is configured to receive logs of the chatbot so that the chatbot's conversation with the user is recorded in real time.

3. The system according to claim 1, wherein the hardware processor is configured to automatically update the training data used to train the chatbot using the correlated potential intents.

4. The system according to claim 1, wherein the hardware processor is configured to automatically start the training of the chatbot.

5. The system according to claim 1, wherein the hardware processor is configured to initiate the training of the chatbot by providing, via a graphical user interface, prompts to add the correlated latent intents to the training data for training the chatbot.

6. The system according to claim 1, wherein the hardware processor is configured to invoke unsupervised machine learning to cluster the chatbot logs into clusters of potential intents and to detect trending topics by analyzing at least social media content in parallel.

7. A computer receives a chatbot log which includes at least the questions received by the chatbot that is conversing with the user, The computer clusters chatbot logs into clusters of potential intents, Computers can detect trending topics by at least analyzing social media content, The computer calculates the semantic similarity between the potential intent and the trend topic, The computer correlates at least one of the potential intents to at least one of the trend topics based on the calculated semantic similarity, A computer implementation method comprising: a computer initiating training of the chatbot using the chatbot's logs related to the correlated potential intents.

8. The method according to claim 7, wherein the chatbot's log is received in real time so that the conversation between the chatbot and the user is recorded in real time.

9. The method according to claim 7, further comprising the computer automatically updating the training data used to train the chatbot using the correlated potential intents.

10. The method according to claim 7, wherein the training of the chatbot is initiated automatically.

11. The method according to claim 7, wherein initiating training of the chatbot includes providing a prompt via a graphical user interface to add the correlated potential intents to training data for training the chatbot.

12. The method according to claim 7, wherein calling unsupervised machine learning to cluster the chatbot logs into clusters of potential intents and detecting trending topics by analyzing at least social media content is performed in parallel.

13. A computer program including program instructions, wherein the program instructions are readable by the device, The chatbot receives a log containing at least the questions received by the chatbot that interacts with the user. The chatbot logs are clustered into clusters of potential intents. At the very least, by analyzing social media content, we can detect trending topics. The semantic similarity between the aforementioned latent intent and the aforementioned trend topic is calculated, Based on the calculated semantic similarity, at least one of the potential intents is correlated with at least one of the trend topics. A computer program that initiates training of the chatbot using the chatbot's logs related to the correlated potential intents.

14. The computer program according to claim 13, wherein the device is configured to receive logs of the chatbot in real time so that the conversation between the user and the chatbot is recorded in real time.

15. The computer program according to claim 13, wherein the device is configured to automatically update the training data used to train the chatbot using the correlated potential intents.

16. The computer program according to claim 13, wherein the device is configured to automatically start the training of the chatbot.

17. The computer program according to claim 13, wherein the device is configured to initiate the training of the chatbot by providing, via a graphical user interface, prompts for adding the correlated potential intents to training data for training the chatbot.

18. The computer program according to claim 13, wherein the device is configured to invoke unsupervised machine learning to cluster the chatbot logs into clusters of potential intents and to detect trending topics by analyzing at least social media content in parallel.

Images