SYSTEMS AND METHODS FOR VEHICLE PURCHASE RECOMMENDATIONS
A vehicle data system may receive, via a website, a user query about a vehicle or features of a vehicle that may not actually exist. The vehicle data system can transform a set of features representing a query vehicle associated with the user query into a query vehicle feature vector, compare the query vehicle feature vector with one or more inventory vehicle feature vectors representative of one or more dealer inventory vehicles, determine a similarity score between the query vehicle and the one or more dealer inventory vehicles, and generate at least one recommendation based on the similarity score, for instance, a great match and/or a good match to the vehicle that the user has inquired about. The at least one recommendation can be presented via the website in real time.
1 . A system, comprising:
one or more computing devices; and
a vehicle data system embodied on at least one server machine communicatively connected to the one or more computing devices, the vehicle data system comprising a processing module, wherein the processing module is configured to:
receive a user query about vehicle features via a website;
transform a set of features representing a query vehicle associated with the user query into a query vehicle feature vector;
compare the query vehicle feature vector with one or more inventory vehicle feature vectors representative of one or more dealer inventory vehicles;
determine a similarity score between the query vehicle and the one or more dealer inventory vehicles;
generate at least one recommendation based on the similarity score; and
present the at least one recommendation via the website.
2 . The system of claim 1 , wherein contribution by a feature of the set of features to the similarity score is determined based on a weight of the feature and a bias function.
3 . The system of claim 1 , further comprising:
normalizing the similarity score.
4 . The system of claim 1 , wherein transformation of the set of features comprises transforming at least one continuous variable associated with the set of features using a scale transformation.
5 . The system of claim 1 , wherein transformation of the set of features comprises transforming at least one discrete variable associated with the set of features using a binary transformation.
6 . The system of claim 1 , wherein transformation of the set of features comprises mapping a preference matrix to a weight of an exclusive feature in the set of features.
7 . The system of claim 6 , wherein the preference matrix comprises a color transition matrix.
8 . A computer program product comprising at least one non-transitory computer readable medium storing instructions translatable by at least one processor of a vehicle data system to perform:
receiving a user query about vehicle features via a website;
transforming a set of features representing a query vehicle associated with the user query into a query vehicle feature vector;
comparing the query vehicle feature vector with one or more inventory vehicle feature vectors representative of one or more dealer inventory vehicles;
determining a similarity score between the query vehicle and the one or more dealer inventory vehicles;
generating at least one recommendation based on the similarity score; and
presenting the at least one recommendation via the website.
9 . The computer program product of claim 8 , wherein contribution by a feature of the set of features to the similarity score is determined based on a weight of the feature and a bias function.
10 . The computer program product of claim 8 , wherein transformation of the set of features comprises transforming at least one continuous variable associated with the set of features using a scale transformation.
11 . The computer program product of claim 8 , wherein transformation of the set of features comprises transforming at least one discrete variable associated with the set of features using a binary transformation.
12 . The computer program product of claim 8 , wherein transformation of the set of features comprises mapping a preference matrix to a weight of an exclusive feature in the set of features.
13 . The computer program product of claim 12 , wherein the preference matrix comprises a color transition matrix.
14 . A method, comprising:
a vehicle data system embodied on at least one server machine receiving a user query about vehicle features via a website;
the vehicle data system transforming a set of features representing a query vehicle associated with the user query into a query vehicle feature vector;
the vehicle data system comparing the query vehicle feature vector with one or more inventory vehicle feature vectors representative of one or more dealer inventory vehicles;
the vehicle data system determining a similarity score between the query vehicle and the one or more dealer inventory vehicles;
the vehicle data system generating at least one recommendation based on the similarity score; and
the vehicle data system presenting the at least one recommendation via the website.
15 . The method according to claim 14 , wherein contribution by a feature of the set of features to the similarity score is determined based on a weight of the feature and a bias function.
16 . The method according to claim 14 , further comprising:
the vehicle data system normalizing the similarity score.
17 . The method according to claim 14 , wherein transformation of the set of features comprises transforming at least one continuous variable associated with the set of features using a scale transformation.
18 . The method according to claim 14 , wherein transformation of the set of features comprises transforming at least one discrete variable associated with the set of features using a binary transformation.
19 . The method according to claim 14 , wherein transformation of the set of features comprises mapping a preference matrix to a weight of an exclusive feature in the set of features.
20 . The method according to claim 19 , wherein the preference matrix comprises a color transition matrix.