Consistent Phrase Relevance Measures

1. Consistent Phrase Relevance Measures Scott Wen-tau Yih & Chris Meek Microsoft Research

2. Why Measure Phase Relevance? • Keyword-driven Online Advertising • Sponsored Search • Ads with bid keywords that match the query • Contextual Advertising (keyword-based) • Ads with bid keywords that are relevant to the content • To deliver relevant ads leads to problems related to phrase relevance measures.

3. queryflight to kyoto Sponsored Search Are these ads relevant to the query?

4. Contextual Advertising How relevant are the keywords behind the ads?

5. Problem – Phrase Relevance Measures • Given a document d and a phrase ph, we want to measure whether ph is relevant to d (e.g., p(ph|d)) Applications – judging ad relevance • Sponsored search (query vs. ad landing page) • Ad relevance verification • Whether a keyword/query is relevant to the page • Contextual advertising (page vs. bid keyword) • External keyword verification • Whether the new keyword is relevant to the content page

6. Keyword Extraction for In-doc Phrases • For in-document phrases, we can use keyword extractor (KEX) directly [Yih et al. WWW-06] • Machine Learning model learned by logistic regression • Use more than 10 categories of features • e.g., position, format, hyperlink, etc. TrueCredit Get immediate access to your complete credit report from 3 credit bureaus. Just $14.95 per month, including $25K ID Theft insurance. Contact TransUnion for more detail… Digital Camera Review The new flagship of Canon’s S-series, PowerShot S80 digital camera, incorporates 8 megapixels for shooting still images and a movie mode that records an impressive 1024 x 768 pixels. What if the phrase is NOT in the document? KEX

7. Challenges of Handling Out-of-doc Phrases • Given a document d and a phrase ph that is not in d • Estimate the probability that ph is relevant to d TrueCredit Get immediate access to your complete credit report from 3 credit bureaus. Just $14.95 per month, including $25K ID Theft insurance. Contact TransUnion for more detail…

8. Challenges of Handling Out-of-doc Phrases • Given a document d and a phrase ph that is not in d • Estimate the probability that ph is relevant to d • Challenges • How do we measure it? • Lack of contextual information that in-doc phrases have • Consistent with the probabilities of in-doc phrases • May need some methods to calibrate probabilities

9. Two Approaches • Calibrated cosine similarity methods • Treat in-doc and out-of-doc phrases equally • Map cosine similarity scores to probabilities • Regression methods based on semantic kernels • Given robust in-doc phrase relevance measures • Predict out-of-doc phrase relevance using similarity between the target phrase and in-doc phrases • Regression methods achieve better empirical results

10. Outline • Introduction • Relevance measures using cosine similarity • Out-of-doc phrase relevance measure using Gaussian process regression • Experiments • Conclusions

11. Similarity-based Measures • Step 1: Estimate sim(d,ph)→ R • Represent das a sparse word vector • Words in document d, associated with weights • Vec(d) = {‘truecredit’,0.9; ‘transunion’,0.7; ‘access’,0.1; … } • Represent phas a sparse word vector via query expansion • Issue ph as a query to search engine; let the result page be document d’ • Vec(ph) ← Vec(d’) • sim(d,ph) = cosine(Vec(d),Vec(ph)) • Choices of term-weighing schemes • Bag of words (SimBin), TFIDF (SimTFIDF) • Keyword Extraction (SimKEX)

12. Map Similarity Scores to Probabilities • Step 2: Map sim(d,ph) to prob(ph|d) • Via a sigmoid function where the weights are pre-learned[Platt ’00] • The sigmoid function can be used to combine multiple relevance scores • SimCombine: Combine SimBin, SimTFIDF & SimKEX

13. Outline • Introduction • Relevance Measures using cosine similarity • Out-of-doc phrase relevance measure using Gaussian process regression • Experiments • Conclusions

14. Regression-basedMeasures:Intuition Relevant in-doc phrases: TrueCredit, TransUnion TrueCredit Get immediate access to your complete credit report from 3 credit bureaus. Just $14.95 per month, including $25K ID Theft insurance. Contact TransUnion… Out-of-doc phrases: credit bureau report vs. Olympics Which out-of-doc phrase is more relevant?

15. Regression-based Measures: Procedure • Step 1: Estimate probabilities of in-doc phrases • KEX(d) = {(‘truecredit’,0.88),(‘transunion’,0.71), (‘credit bureaus’,0.64), (‘id theft’,0.14)} • Step 2: Represent each phrase as a TFIDF vector via query expansion • x1=Vec(‘truecredit’), y1=0.88; x2=Vec(‘transunion’), y2=0.71x3=Vec(‘credit bureaus’), y3=0.64; x4=Vec(‘id theft’), y4=0.14 • Step 3: Represent the target phrase ph as a vector • x=Vec(ph), y=? • Step 4: Use a regression model to predict y • Input: (x1, y1), …, (xn, yn) and x • Output: y

16. Gaussian Process Regression (GPR) • We don’t specify the functional form of the regression model • Instead, we only need to specify the “kernel function” • k(x1,x2): linear kernel, polynomial kernel, RBF kernel, etc. • Conceptually, kernel function tells how similar x1 & x2 are • Changing kernel function changes the regression function • Linear kernel → Bayesian linear regression (x1,y1), (x2,y2),…, (xn,yn) GPR y x O(N3) from matrix inversion, where N≤20 typically kernel function e.g., k(xi,xj) = xi·xj

17. Outline • Introduction • Relevance Measures using cosine similarity • Out-of-doc phrase relevance measure using Gaussian process regression • Experiments • Conclusions

18. Data • From sponsored search ad-click logs (3-month period in 2007) • Randomly select 867 English ad landing pages • Each page is associated with the original query and ~10 related keywords (from internal query suggestion algorithms) • Labeled 9,319 document-keyword pairs • 4,381 (47%) relevant; 4,938 (53%) irrelevant • Most keywords (81.9%) are out-of-document • 10-fold cross-validation when learning is used

19. Evaluation Metrics • Accuracy • Quality of binary classification • False positive and false negative are treated equally • AUC (Area Under the ROC curve) • Quality of ranking • Equivalent to pair-wise accuracy • Cross Entropy • Quality of probability estimations • -log2[p(ph|d)] if ph is labeled relevant to d • -log2[1-p(ph|d)] if ph is labeled irrelevant to d

20. Accuracy Better

21. AUC Scores Better

22. Cross Entropy Better

23. Conclusions (1/2) • Phrase relevance measure is a crucial task for online advertising • Our solution: similarity & regression based methods • Consistent probabilities for out-of-doc phrases • Similarity-based methods • Simple and straightforward • The combined approach can lead to decent performance • Regression-based methods • Achieved the best results in our experiments • Quality depends on the in-doc relevance estimates & kernel

24. Conclusions (2/2) • Future Work – More machine learning techniques • SimCombine • An ML method using basic similarity measures as features • Explore more features (e.g., query frequency, page quality) • Other machine learning models • Gaussian process regression • Learning a better kernel function • Kernel meta-training [Platt et al. NIPS-14] • Maximum likelihood training