A system for measuring semantic relevance of keywords by parsing information resources, comprising:
- a data processing system having one or more processors, configured to:
- receive, from a content provider computing device corresponding to a domain entity, a seed keyword to generate additional keywords relevant to the seed keyword;
- access a semantic relationship graph including a plurality of nodes and a plurality of edges, each of the plurality of nodes referencing at least one corresponding keyword, each of the plurality of edges connecting a corresponding pair of the plurality of nodes and defining a corresponding semantic distance between a pair of at least one keywords referenced by the pair of the plurality of nodes;
- identify, using a semantic relationship graph, a plurality of keyword categories including a first keyword category and a second keyword category, each of the plurality of keyword categories including a plurality of keywords, each of the plurality of keywords having a semantic distance from the seed keyword less than a semantic distance threshold as defined by a corresponding plurality of edges in the semantic relationship graph;
- identify an information resource of the domain entity using the web address, the information resource having primary content;
- determine, for each keyword of the first keyword category and the second keyword category, a frequency of the keyword occurring with the seed keyword on the primary content of information resource of the domain entity identified using the web address;
- determine, for each keyword of the first keyword category and the second keyword category, a keyword-seed affinity score based on the frequency of the keyword occurring with the seed keyword on the primary content of the information resource of the domain entity identified using the web address;
- determine, for the first keyword category, a first category-seed affinity score based on the keyword-seed affinity scores for each of the plurality of keywords in the first keyword category;
- determine, for the second keyword category, a second category-seed affinity score based on the keyword-seed affinity scores for each of the plurality of keywords in the second keyword category;
- identify a semantic conflict between the first keyword category and the second keyword category based on a semantic distance between at least one of the plurality of keywords in the first keyword category and at least one of the plurality of keywords in the second keyword category as defined by a corresponding edge within the semantic graph;
- compare, responsive to identifying the semantic conflict between the first keyword category and the second keyword category, the first category-seed affinity score to the second category-affinity score;
- and
- determine, based on the comparison, that the first category-seed affinity score is greater than the second category-seed affinity score;
- and
- transmit, to a content provider computing device for display, the plurality of keywords of the first keyword category and the plurality of keywords of the second keyword category, the plurality of keywords in the first keyword category indicated as selected responsive to determining that the first category-seed affinity score is greater than the second category-seed affinity score, the plurality of keywords in the second keyword category indicated as unselected responsive to determining that the second category-seed affinity score is less than the first category-seed affinity score.