Ad bidding based on a buyer-defined function
An ad processor evaluates bid functions that are based on concepts that might be generated from interpretations of natural language expressions. Ad buyers provide the functions with corresponding ads to ad processors. Bid functions are further based on the values of semantic information referenced by expressions. Bid functions are further based on environmental information. Ad buyers are able to modify bid functions. Ads may be provided in the form of questions, and may be indicated by an identifying sound. Upon finding no expression concepts within a bid function, the set of expression concepts is expanded according to strengths of connections between concepts in a concept graph.
1. A method of selecting one of a plurality of ads, the method comprising:
providing a concept graph including a plurality of nodes and a plurality of connections, each node of the plurality of nodes corresponding to a concept and each connection connecting a pair of nodes of the plurality of nodes and having a weight indicating a strength of a relationship between the pair of nodes, the weight being generated by evaluating a corpus of text and determining the strength of the relationship between the pair of nodes according to at least one of (a) proximity of occurrences of the concepts to which the pair of nodes correspond and (b) occurrence of the concepts to which the pair of nodes correspond in definitions of one another in a dictionary;
performing by a computer system including one or more processing devices:
interpreting a natural language expression;
reading, from a data source, a value of semantic information indicated by the expression, the data source being a dynamic data source;
reading, from a database of ads, a first bid function for a first ad and a second bid function for a second ad, each bid function of the first bid function and second bid function being applicable to the semantic information;
applying the first bid function to the value of the semantic information to generate a first bid amount, the first bid function referencing one or more first concepts and being based on one or more weights of one or more connections of the plurality of connections in the concept graph between the one or more first concepts and one or more concepts corresponding to the natural language expression;
applying the second bid function to the value of the semantic information to generate a second bid amount, the second bid function referencing one or more second concepts and being based on one or more weights of one or more connections of the plurality of connections in the concept graph between the one or more second concepts and the one or more concepts corresponding to the natural language expression;
comparing the first bid amount to the second bid amount;
determining that the first bid amount is larger than the second bid amount; and
in response to determining that the first bid amount is larger than the second bid amount, selecting the first ad;
wherein at least one node of the plurality of nodes further includes a plurality of synonyms corresponding to the concept to which the at least one node corresponds.
2. The method of claim 1 , further comprising applying each bid function of the first and second bid functions to the semantic information by calling a library function.
3. The method of claim 1 , further comprising selecting content of the first ad based on the value of the semantic information.
4. The method of claim 1 , wherein the semantic information comprises a plurality of values.
5. The method of claim 1 , wherein the semantic information is about a weather condition.
6. The method of claim 1 , wherein the semantic information is about a geographic location.
7. The method of claim 1 , wherein the semantic information is about a person.
8. The method of claim 1 , wherein the semantic information is about a time.
9. The method of claim 1 , wherein the semantic information is about a price.
10. The method of claim 1 , further comprising weighting the first bid amount based on an environmental overlay such that where the one or more concepts defined in the first bid function are weighted according to relationship to at least one of a current geographic location of a user, current time, gender of a user, and proximity of the user to the user's home.
11. The method of claim 1 , further comprising:
providing a natural language response; and
providing the selected ad content with the natural language response to a source of the natural language expression.
12. The method of claim 1 , further comprising receiving a modification of the first bid function;
in response to receiving the modification of the first bid function, modifying the first bid function without modifying the first ad.
13. The method of claim 1 , further comprising:
reading information from an ad buyer;
defining the first bid function based on the information from the ad buyer information.
14. The method of claim 1 , further comprising:
reading information from a third-party data source;
defining the first bid function based on the information from the third-party data source.