Method and system for automatically prioritizing content provided to a user

1. A method for interpreting user-specific text using a model for natural language processing and optimally personalizing content, comprising:

receiving a plurality of user-specific text strings associated with a user;

using the model for natural language processing, the model comprising a trained semantic embedding model, generating a set of user-specific embeddings, comprising, for each user-specific text string of the plurality: generating a respective user-specific embedding based on the user-specific text string, wherein the respective user-specific embedding preserves semantic language information from the user-specific text string;

generating a set of content embeddings, comprising, for each content block of a set of content blocks, generating a respective content embedding associated with the content block;

based on a set of semantic similarity metrics determined between the set of user-specific embeddings and the set of content embeddings, determining a ranked list of content blocks selected from the set of content blocks, wherein determining the ranked list is performed in response to receiving the set of user-specific embeddings;

in response to determining the ranked list, storing the ranked list in a database;

after storing the ranked list, receiving a content request from the user;

in response to receiving the content request, retrieving the ranked list and selecting the at least one content block based on the ranked list;

adjusting the ranked list to exclude a content block previously provided to the user;

based on the ranked list, providing at least one content block to the user, wherein providing the at least one content block to the user is performed in response to selecting the at least one content block;

after providing the at least one content block to the user:

receiving a second plurality of user-specific text strings associated with the user;

using the trained semantic embedding model, generating a second set of user-specific embeddings, comprising, for each user-specific text string of the second plurality: generating a respective user-specific embedding based on the user-specific text string;

based on the second set of user-specific embeddings and a second set of content embeddings, determining a second ranked list of content blocks selected from the set of content blocks;

in response to determining the second ranked list, storing the second ranked list;

after storing the second ranked list, receiving a second content request from the user;

in response to receiving the second content request:

retrieving the second ranked list;

selecting a second content block based on the second ranked list; and

providing the second content block to the user.

2. The method of claim 1 , wherein:

the set of user-specific embeddings are defined in an embedding space; and

the set of content embeddings are defined in the embedding space.

3. The method of claim 2 , wherein generating the set of content embeddings is performed using the trained semantic embedding model.

4. The method of claim 3 , wherein the trained semantic embedding model comprises a multi-lingual semantic embedding model.

5. The method of claim 2 , wherein determining the ranked list comprises:

for each content embedding of the set of content embeddings:

determining a respective set of similarity metrics, wherein the set of semantic similarity metrics comprises the respective set of similarity metrics, wherein each similarity metric is indicative of similarity of the content embedding to a respective user-specific embedding of the set of user-specific embeddings; and

determining an overall relevance metric associated with the content embedding; and

ranking the content embeddings of the set based on the overall relevance metrics.

6. The method of claim 5 , wherein each similarity metric is determined based on a cosine similarity between a content embedding and a user-specific embedding.

7. The method of claim 5 , wherein, for each content embedding of the set: determining the overall relevance metric comprises selecting the highest similarity metric of the respective set of similarity metrics.

8. The method of claim 1 , further comprising generating the second set of content embeddings, comprising, for each content block of a second set of content blocks, generating a respective content embedding associated with the content block.

9. The method of claim 1 , wherein the second set of content embeddings is equivalent to the set of content embeddings.

10. The method of claim 1 , wherein:

receiving the plurality of user-specific text strings comprises:

receiving a set of messages of a first conversational session with the user; and

selecting the plurality of user-specific text strings from the set of messages; and

receiving the second plurality of user-specific text strings comprises:

receiving a summary of a second conversational session with the user; and

selecting the second plurality of user-specific text strings from the summary.

11. The method of claim 1 , wherein receiving the plurality of user-specific text strings comprises:

receiving a summary of a conversational session with the user; and

selecting a set of user-specific text strings from the summary, wherein the plurality of user-specific text strings comprises the selected set of user-specific text strings.

12. The method of claim 1 , wherein receiving the plurality of user-specific text strings comprises:

receiving a set of messages of a conversational session with the user; and

selecting a set of user-specific text strings from the set of messages, wherein the plurality of user-specific text strings comprises the selected set of user-specific text strings.

13. The method of claim 12 , wherein selecting the set of user-specific text strings from the set of messages comprises, using a trained importance model, selecting a subset of messages from the set of messages, wherein each user-specific text string comprises a different message of the subset.

14. The method of claim 1 , wherein providing the at least one content block to the user comprises:

selecting a plurality of content blocks from the ranked list; and

displaying the plurality of content blocks to the user in order of ranking in the ranked list.