Computer implemented system and method for enhancing keyword expansion
View Patent ↗A computerized system which implements a method for obtaining an improved set of relevant keywords from an initial set of relevant keywords and their associated relative ratings, utilizing the proven expertise and reliability of existing keyword expansion tools and the accumulated experience of search engines. The method comprises iterating over keyword seeds, search engines' results and keyword suggestion sources' results, merging lists of keywords and combining their scores by means of an optimization method and learned effectiveness of result sources. Keyword relevance scores may be effectively used to distribute a content provider's resources amongst keywords thereby to achieve inclusion and optimization in search engines' results. The system input is a set of one or more seed keywords and the system produces an expanded list of unique keywords, each keyword typically accompanied by a computed relevance score.
1. A computerized method for enhancing keyword expansion for promoting digital content, comprising:
accessing a plurality of search engines using at least one input seed keyword, each of the plurality of search engines returns organic results respective of the at least one input seed keyword;
for each organic result returned by each of the plurality of search engines, retrieving, using a plurality of keyword suggestion sources, a list of key terms and their scores;
merging the list of key terms and their scores from each of the plurality of keyword suggestion sources into a merged list respective of each of the at least one input seed keyword and scores thereof;
combining the scores for each of the key terms in the merged list into a single combined score;
associating the single combined score with a respective input seed keyword; and
reporting the single combined score for each of the key terms for the at least one input seed keyword and the merged list resulting from access of the plurality of search engines and keyword suggestion sources, wherein the merged list contains the keyword expansion.
2. The computerized method according to claim 1 , further comprises:
selecting said key terms based on a popularity criterion.
3. The computerized method according to claim 1 , further comprises:
selecting said key terms based on a criterion of relatedness.
4. The computerized method according to claim 1 , wherein enhancing keyword expansion further comprises: iterating a plurality of result pages.
5. The computerized method according to claim 4 , further comprises: reusing results as secondary seeds for further refinement.
6. The computerized method according to claim 1 , wherein enhancing keyword expansion further comprises: iterating a plurality of seed keywords.
7. The computerized method according to claim 6 , further comprises: reusing results as secondary seeds for further refinement.
8. The computerized method according to claim 1 , further comprises:
merging results within an individual seed keyword using the at least one input seed keyword.
9. The computerized method according to claim 8 , further comprises reusing results as secondary seeds for further refinement.
10. The computerized method according to claim 1 , further comprises: merging results between a plurality of seed keywords.
11. The computerized method according to claim 10 , further comprises: reusing results as secondary seeds for further refinement.
12. The computerized method according to claim 1 , further comprises: combining scores is performed using an optimization method.
13. The computerized method according to claim 12 , further comprises: reusing results as secondary seeds for further refinement.
14. The computerized method according to claim 12 , wherein combing the scores further comprising:
generating a combined score of a key term by summing the multiplication of the scores of the respective key terms retrieved from the keyword suggestion sources with weights, wherein the weights are constrained to be non-negative and the constraint is solved using the optimization method.
15. The computerized method according to claim 1 , further comprises: reusing results as secondary seeds for further refinement.
16. The computerized method according to claim 1 , further comprises: reuse reusing results as secondary seeds for further refinement.
17. The computerized method according to claim 1 , further comprising: selecting using a cut-of-rule a set of organic results out of the retrieved organic results to be processed.