Offering automobile recommendations from generic features learned from natural language inputs

1. An apparatus, comprising:

a memory to store instructions; and

processing circuitry, coupled with the memory, operable to execute the instructions, that when executed, cause the processing circuitry to:

receive a natural language automobile request from a user device, the request initiated by a user and including generic language with respect to at least one automobile characteristic;

apply a machine learning model (MLM) to the received request, wherein the MLM is trained, using a data set based on a corpus of one or more automobile reviews, to determine an association between one or more specific automobile makes and models referenced in the corpus of one or more automobile reviews and generic language referenced in the corpus of one or more automobile reviews, wherein the data set includes a plurality of sentences from the corpus of one or more automobile reviews;

generate, by the MLM based on the request, a recommendation of at least one specific automobile feature, wherein the at least one specific automobile feature is determined from a ranking of a plurality of automobile makes and models, each one of the plurality of automobile makes and models associated with a confidence value based on the generic language with respect to the at least one automobile characteristic, wherein the MLM is further trained by pre-processing the data set to remove predetermined verbs, pronouns, and stop words from the plurality of sentences;

populate an interface with at least one make and model for the user based on the ranking of the plurality of automobile makes and models; and

update the MLM based on an adjustment to a probability distribution as a result of a user provided preference for the at least one make and model.

2. The apparatus of claim 1 , the processing circuitry being further caused to: adjust the ranking based on at least one additional feature, including at least one of i) a price range provided by the request and ii) a preferred location for interacting with a selected automobile.

3. The apparatus of claim 1 , wherein the one or more automobile reviews are one or more expert reviews, wherein the data set is based on i) generic language related to a plurality of automobile makes and models associated with the corpus of one or more automobile reviews and ii) specific language related to at least one feature of at least one of the plurality of automobile makes and models associated with the corpus of one or more automobile reviews.

4. The apparatus of claim 3 , wherein the MLM is trained to determine an association of a specific automobile make and model referenced in the corpus of one or more automobile reviews to the generic language of the corpus of one or more automobile reviews by analyzing a relationship between the generic language of the one or more automobile reviews and the specific language of the one or more automobile reviews.

5. The apparatus of claim 1 , wherein the MLM is an embedded MLM, wherein the data set includes a plurality of sentences from the corpus of one or more automobile reviews, and wherein the pre-processed data is processed by a sentence encoder that is part of the MLM, the apparatus further comprising:

a second interface configured to receive the request, the second interface including either i) a chatbot or ii) a single field for submitting an entirety of the request, and the processing circuitry to generate the recommendation further comprising instructing a display to display the recommendation.

6. A non-transitory computer-readable storage medium storing computer-readable program code executable by a processor to:

receive a natural language automobile request from a user device, the request including generic language with respect to at least one automobile feature;

apply a word-frequency based machine learning model (MLM) to the received request, wherein the MLM is trained, using a data set based on a corpus of one or more automobile reviews, to determine an association between one or more specific automobile makes and models referenced in the corpus of one or more automobile reviews and generic language referenced in the corpus of one or more automobile reviews; and

generate, by the MLM, a recommendation for at least one specific automobile feature based on the request, and

wherein the data set includes a plurality of sentences from the corpus of one or more automobile reviews, wherein the MLM training includes: pre-processing the data set, by a sentence encoder, to remove predetermined verbs, pronouns, and stop words from the plurality of sentences, and

wherein the determined association between the one or more specific automobile makes and models to the generic language of the corpus of one or more automobile reviews is based on a frequency of both one type of generic language of the data set and a specific functional language related to at least one vehicular feature being included in the data set.

7. The non-transitory computer-readable storage medium of claim 6 , wherein the corpus of one or more reviews is a corpus of one or more expert reviews.

8. The non-transitory computer-readable storage medium of claim 6 , further comprising computer-readable program code executable to:

generate a user interface on a computer display, the interface including a single field for entering an entirety of the request; and

cause the computer display to display the recommendation, wherein the request consists solely of generic language.

9. The non-transitory computer-readable storage medium of claim 6 , wherein the MLM is an embedded MLM, wherein the request is received via one of a chatbot or a single input field for entering an entirety of the request.

10. A method comprising:

receiving a natural language automobile request from a user device, the request including generic language with respect to at least one automobile characteristic;

applying a machine learning model (MLM) to the received request, wherein the MLM is trained, using a data set based on a corpus of one or more automobile reviews, to determine an association between one or more specific automobile makes and models referenced in the corpus of one or more automobile reviews and generic language referenced in the corpus of one or more automobile reviews, wherein the data set includes a plurality of sentences from the corpus of one or more automobile reviews;

generating, by the MLM responsive to the request, a recommendation of at least one specific automobile feature, wherein the at least one specific automobile feature is determined from a ranking of a plurality of automobile makes and models, each one of the plurality of automobile makes and models associated with a confidence value based on the generic language with respect to the at least one automobile characteristic;

pre-processing, by a sentence encoder, the data set to remove predetermined verbs, pronouns, and stop words from the plurality of sentences;

populating an interface with at least one make and model based on the ranking of the plurality of automobile makes and models; and

updating the MLM based on an adjustment to a probability distribution as a result of a user provided preference for the at least one make and model.

11. The method of claim 10 , the method further comprising:

adjusting the ranking based on at least one additional feature, including at least one of i) a price range provided by the request and ii) a preferred location for interacting with a selected automobile.

12. The method of claim 11 , wherein the corpus of one or more automobile reviews is a corpus of one or more expert automobile reviews, wherein the data set is based on i) generic language related to a plurality of automobile makes and models associated with the corpus of one or more expert automobile reviews and ii) specific language related to at least one feature of at least one of the plurality of automobile makes and models associated with the corpus of one or more expert automobile reviews, and wherein the generic language of the one or more expert automobile reviews is related to the specific language of the one or more expert automobile reviews.

13. The method of claim 10 , wherein the MLM is an embedded MLM, wherein the request is received via one of a chatbot or a single input field for entering an entirety of the request.