Prediction-model-based mapping and/or search using a multi-data-type vector space

Abstract

In certain embodiments, content items may be obtained, where each of the content items may include multiple data types. Machine learning models may be caused to be trained based on the content items to map data in a vector space by providing at least a first portion of each of the content items as input to at least one of the machine learning models and providing at least a second portion of each of the content items as input to at least another one of the machine learning models. A search request for results may be obtained, where the search request includes search parameters. One or more locations within the vector space may be predicted (e.g., by one or more of the machine learning models) based on the search parameters. Information (indicating content items mapped to or proximate the predicted locations) may be provided as a request response.

What is claimed is: 
     
         1 . An apparatus comprising:
 a network interface configured to receive a search request for a content item that includes a plurality of data types, where the search request includes one or more search parameters; and   a processor configured to
 predict a region within vector space that satisfies the search request based on the one or more search parameters, 
 identify a plurality of different distance thresholds of the region within the vector space which are assigned to the plurality of different data types in a plurality of different dimensions within the vector space, respectively, 
 identify search results that satisfy the plurality of different distance thresholds of the region of the vector space and output the search results via a user interface. 
   
     
     
         2 . The apparatus of  claim 1 , wherein the processor is configured to map a content item of the search result into a plurality of vectors in vector space and compare locations of the plurality of vectors within the vector space to the plurality of different distance thresholds within the vector space. 
     
     
         3 . The apparatus of  claim 2 , wherein the processor is configured to map a plurality of different data types of the content item of the search result into vector space via a plurality of machine learning models for the plurality of different data types, respectively. 
     
     
         4 . The apparatus of  claim 2 , wherein the processor is configured to generate a single resulting vector from the plurality of vectors in vector space, and compare the single resulting vector to the plurality of different distance thresholds within the vector space. 
     
     
         5 . The apparatus of  claim 1 , wherein the one or more search parameters comprise at least one of a keyword, an image, an audio, a logical operator, and a pointer to a web resource of the content item. 
     
     
         6 . The apparatus of  claim 1 , wherein the processor is configured to predict the region within the vector space based on a machine learning model which receives the one or more search parameters as input and outputs the region. 
     
     
         7 . The apparatus of  claim 1 , wherein the processor is further configured to train a plurality of machine learning models to map content items to the vector space based on the identified search results. 
     
     
         8 . The apparatus of  claim 7 , wherein the processor is configured to train the plurality of machine learning models via a second and separate instance of the vector space. 
     
     
         9 . A method comprising:
 receiving a search request for a content item that includes a plurality of data types, where the search request includes one or more search parameters;   predicting a region within vector space that will satisfy the search request based on the one or more search parameters;   identifying a plurality of different distance thresholds of the region within the vector space which are assigned to the plurality of different data types in a plurality of different dimensions within the vector space, respectively; and   identifying search results that satisfy the plurality of different distance thresholds of the region of the vector space, and returning the search results via a user interface.   
     
     
         10 . The method of  claim 9 , wherein the identifying comprises mapping a content item of the search result into a plurality of vectors in vector space and comparing locations of the plurality of vectors within the vector space to the plurality of different distance thresholds within the vector space. 
     
     
         11 . The method of  claim 10 , wherein the mapping comprises mapping a plurality of different data types of the content item of the search result into vector space via a plurality of machine learning models for the plurality of different data types, respectively. 
     
     
         12 . The method of  claim 10 , wherein the mapping further comprises generating a single resulting vector from the plurality of vectors in vector space, and comparing the single resulting vector to the plurality of different distance thresholds within the vector space. 
     
     
         13 . The method of  claim 9 , wherein the one or more search parameters comprise at least one of a keyword, an image, an audio, a logical operator, and a pointer to a web resource of the content item. 
     
     
         14 . The method of  claim 9 , wherein the predicting comprises predicting the region within the vector space based on a machine learning model which receives the one or more search parameters as input and outputs the region. 
     
     
         15 . The method of  claim 9 , wherein the method further comprises training a plurality of machine learning models to map content items to the vector space based on the identified search results. 
     
     
         16 . The method of  claim 15 , wherein the training of the plurality of machine learning models is performed in a second and separate instance of the vector space. 
     
     
         17 . A non-transitory computer-readable medium comprising instructions which when executed by a processor cause a computer to perform a method comprising:
 receiving a search request for a content item that includes a plurality of data types, where the search request includes one or more search parameters;   predicting a region within vector space that will satisfy the search request based on the one or more search parameters;   identifying a plurality of different distance thresholds of the region within the vector space which are assigned to the plurality of different data types in a plurality of different dimensions within the vector space, respectively; and   identifying search results that satisfy the plurality of different distance thresholds of the region of the vector space, and returning the search results via a user interface.   
     
     
         18 . The non-transitory computer-readable medium of  claim 17 , wherein the identifying comprises mapping a content item of the search result into a plurality of vectors in vector space and comparing locations of the plurality of vectors within the vector space to the plurality of different distance thresholds within the vector space. 
     
     
         19 . The non-transitory computer-readable medium of  claim 17 , wherein the mapping comprises mapping a plurality of different data types of the content item of the search result into vector space via a plurality of machine learning models for the plurality of different data types, respectively. 
     
     
         20 . The non-transitory computer-readable medium of  claim 17 , wherein the mapping further comprises generating a single resulting vector from the plurality of vectors in vector space, and comparing the single resulting vector to the plurality of different distance thresholds within the vector space.