Presentation is loading. Please wait.

Presentation is loading. Please wait.

18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University1 Keyword Generation for Search Engine Advertising Amruta Joshi*, Yahoo! Research.

Similar presentations


Presentation on theme: "18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University1 Keyword Generation for Search Engine Advertising Amruta Joshi*, Yahoo! Research."— Presentation transcript:

1 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University1 Keyword Generation for Search Engine Advertising Amruta Joshi*, Yahoo! Research Rajeev Motwani, Stanford University * This work was done at Stanford

2 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 2 Search Results Sponsore d Search Results

3 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 3 Long Tail Queries Frequency in query-logs Expensive, high frequency keywords Target inexpensive, low frequency keywords instead

4 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 4 Keyword Pricing

5 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 5 Pick the right keywords Advantages  more focused audience  lesser competition, easier to get #1 position  cost-effective alternative Keywords should be  Highly Relevant to base query  Nonobviousness to guess from the base query E.g.:  hawaii vacation $3  kona holidays $0.11

6 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 6 Objective To generate, with good precision and recall, a large number of keywords that are relevant to the input word, yet non- obvious in nature.

7 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 7 Who’s doing all this? Large Advertisers SEO companies and small start-ups manage advertising profiles Eg: Eventually every advertiser is interested in optimizing his portfolio

8 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 8 Other Techniques … Meta-tag Spidering:  Extract Keyword & Description tags from top search hits  Example of meta-tags for query ‘hawaii travel’ Relevant: hawaii travel, hawaii vacation, hawaiian islands, hawaii tourism Off-topic: hawaii homes, moving to hawaii, hawaii living, hawaii news, living in hawaii, hawaii products, Irrelevant: sovereignty, volcanoes, sports, music

9 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 9 Other Techniques … Proximity-based tools  Pick phrases in the proximity of given word  e.g.: family hawaii vacations, discount hawaii vacations Query log Mining  Suggest popular queries containing seed keywords

10 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 10 Other Techniques Advertiser log mining or Query Co- occurrence based mining  Exploits co-occurrence in advertiser keyword search logs  Increase competition!

11 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 11 Directed Relevance Relationships Word A strongly suggests word B, but the reverse may not hold true AB x BA y x ≠ y railwayseurail 25 railways eurail 2 Example:

12 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 12 Building Context Characteristic Document  Build context of the term using terms found in the proximity of seed term in the top 50 hits from search engine for that term europe. C Search Engine

13 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 13 Building the Graph TermsNet  Nodes = terms  Edges = directed relevance relationships  Weights = strength of directed relationship, i.e., the frequency of destination term in characteristic document of source term

14 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 14 TermsNet europe. C railways C euro C eurail C maps C atlas C schengen C

15 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 15 Ranking Suggestions Quality Score Incorporates  Edge-weights  Normalization for common words Quality Q(x, q) = w x,q / (1+log (1+∑w x,i )) where each i is an outneighbor of ‘x’ xq w x,q

16 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 16 Ratings Relevance  Indicates Relevance of suggested keyword to seed word  Given by human editors  e.g.: For query ‘flights’ Relevance (‘flights’, ‘cathay pacific’) = 1 Relevance (‘flights’, ‘cheap flight’) = 1 Relevance (‘flights’, ‘magazines’) = 0 Nonobviousness  Indicates nonobviousness of suggested keyword relative to seed word  Calculated as:  If No base query word/stem present in suggested keyword, Nonobviousness = 1, else = 0  e.g.: For query ‘flights’ Relevance (‘flights’, ‘cathay pacific’) = 1 Relevance (‘flights’, ‘cheap flight’) = 0 Relevance (‘flights’, ‘magazines’) = 1  Used standard Porter stemmer for automating this rating

17 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 17 Evaluation Evaluation Measures  Average Precision: Ratio of number of relevant keywords retrieved to number of keywords retrieved. Indicates quality of results  Average Recall The proportion of relevant keywords that are retrieved, out of all relevant keywords available. For our expts Recall (T i ) = # retrieved by T i / # retrieved by (T 1 U T 2 U…U T n )  Average Nonobviousness Average of all nonobviousness ratings of suggested keywords

18 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 18 Output for query ‘flights’ Co-occurrence Based Query LogMeta-Tag Spidering Meta-Crawler Lists Query-log Mining TermsNet Airfare airfares airlines Cyprus goa flys holidays trains aer aeroflot aeromexico aircanada alicante bwia heathrow icelandair bookings Consolidator Flights cheap flights airline flights cheap airline flights cheap international flights flights to europe business class flights flights new york australia flights cheap flights to europe cheap flights to orlando cheap flights las vegas track flights flights florida flights europe las flights cheap flights to australia real time flight arrivals airfare flights flight map delays cruises us flight arrivals flight arrivals state map flight arrival flight cancellation s arrival times arrival delays flight departure vacation packages street map air travel airline discount tickets airline fares airline tickets airline tickets under 100 american airlines bargain flights bmibaby british airways british airways flights british airways home page british airways timetable british midland budget airline flight cheap flight las vegas flight flight tracker flight to orlando flight to london flight to new york airline flight flight to los angeles flight 93 flight to fort lauderdale light of the phoenix flight to honolulu flight to chicago flight to miami cheap flights airline flights air newzealand flight prices bmibaby globespan low cost airlines united airlines airline- consolidators charter flights airfare flight reservations cathay pacific british midland airways discount airfare flight tickets jet2 travelocity

19 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 19 Avg. Precision, Recall, Nonobviousness

20 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 20 Evaluation Measures F-measures  Measure of overall performance Harmonic mean of  F(PR) – Avg. Precision & Avg. Recall  F(RN) – Avg. Recall & Avg. Nonobviousness  F(PN) – Avg. Precision & Avg. Nonobviousness  F(PRN) – Avg. Precision, Avg. Recall & Avg. Nonobviousness

21 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 21 F-Measures

22 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 22 Quality of Suggestions over different intervals of ranked results Figure 2: Quality of keywords over different ranked intervals

23 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 23 Future Directions Incorporate keyword frequency in ranking suggestions Incorporate keyword pricing information in ranking suggestions Applications to other domains  Find related movies, papers, people

24 18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University 24 Thank You! Questions?


Download ppt "18 December 2006 Amruta Joshi and Rajeev Motwani, Stanford University1 Keyword Generation for Search Engine Advertising Amruta Joshi*, Yahoo! Research."

Similar presentations


Ads by Google