1 / 18

A Hierarchy of Independence Assumptions for Multi-Relational Bayes Net Classifiers

School of Computing Science Simon Fraser University Vancouver, Canada. A Hierarchy of Independence Assumptions for Multi-Relational Bayes Net Classifiers. Outline. Multi- R elational Classifiers Multi-Relational Independence Assumptions Classification Formulas Bayes Nets Evaluation.

scout
Download Presentation

A Hierarchy of Independence Assumptions for Multi-Relational Bayes Net Classifiers

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. School of Computing Science • Simon Fraser University • Vancouver, Canada A Hierarchy of Independence Assumptions for Multi-Relational Bayes Net Classifiers

  2. Outline • Multi-Relational Classifiers • Multi-Relational Independence Assumptions • Classification Formulas • Bayes Nets • Evaluation A Hierarchy of Independence Assumptions

  3. Database Tables • Tables for Entities, Relationships • Can visualize as network Ranking = 1 Diff = 1 Jack 101 Registration Link-based ClassificationTarget table: Student Target entity: Jack Target attribute (class): Intelligence A Hierarchy of Independence Assumptions

  4. Extended Database Tables A Hierarchy of Independence Assumptions

  5. Multi-Relational Classifiers Count relational features Aggregate relational features • Propositionalization • Example: use average grade • Disadvantages: • loses information • slow to learn (up to several CPU days) Log-Linear Models Example: use number of A,s number of Bs,… ln(P(class)) = Σ xiwi – Z Disadvantage: slow learning + Independence Assumptions • Log-Linear Models With Independencies • Fast to learn • −Independence Assumptions may be only approximately true A Hierarchy of Independence Assumptions

  6. Independence Assumptions A Hierarchy of Independence Assumptions

  7. Independence Assumptions: Naïve Bayes Naive Bayes: non-class attributes are independent of each other, given the target class label. Legend: Given the blue information, the yellow columns are independent. A Hierarchy of Independence Assumptions

  8. Path Independence Naive Bayes: non-class attributes are independent of each other, given the target class label. Legend: Given the blue information, the yellow rows are independent. Path Independence: Links/paths are independent of each other, given the attributes of the linked entities.

  9. Influence Independence Legend: Given the blue information, the yellow columns are independent from the orange columns Influence Independence: Attributes of the target entity are independent of attributes of related entities, given the target class label. Naive Bayes: non-class attributes are independent of each other, given the target class label. Path Independence: Links/paths are independent of each other, given the attributes of the linked entities. Path-Class Independence: the existence of a link/path is independent of the class label.

  10. Classification Formulas • Can rigorously derive log-linear prediction formulas from independence assumptions. • Path Independence: predict max class for: log(P(class|target attributes)) + sum over each table, each row: [log(P(class|information in row)) – log(P(class|targetatts))] • PI + Influence Independence:predict max class for: log(P(class|target attributes)) + sum over each table, each row: [log(P(class|information in row)) – log(prior P(class))] A Hierarchy of Independence Assumptions

  11. Relationship to Previous Formulas A Hierarchy of Independence Assumptions

  12. Evaluation A Hierarchy of Independence Assumptions

  13. Data Sets and Base Classifier • Standard DatabasesKDD Cup, UC Irvine • MovieLens not shown. Country Continent Hepatitis In-Hosp Economy Out-Hosp Mondial Biopsy Patient Government Interferon Borders • Classifier • Can plug in any single-table probabilistic base classifier with classification formula . • We use Bayes nets. Country2 Transaction Loan Account Financial Order Disposition District Card Client

  14. Family of Alarm Burglary Earthquake E B P(A | E,B) e b 0.9 0.1 e b 0.2 0.8 Radio Alarm e b 0.9 0.1 0.01 0.99 e b Call What is a Bayes net? Compact representation of joint probability distributions via conditional independence • Qualitative part: • Directed acyclic graph (DAG) • Nodes - random vars. • Edges - direct influence Quantitative part: Set of conditional probability distributions Together: Define a unique distribution in a factored form Figure from N. Friedman

  15. Independence-Based Learning is Fast strongest assumption weakest assumption Training Time in seconds A Hierarchy of Independence Assumptions

  16. Independence-Based Models are Accurate strongest assumption weakest assumption • Similar results for F-measure, Area Under Curve A Hierarchy of Independence Assumptions

  17. Conclusion • Several plausible independence assumptions/classification formulas investigated in previous work. • Organized in unifying hierarchy. • New assumption: multi-relational path independence. • most general, implicit in other models. • Big advantage: Fast scalable simple learning. • Plug in single-table probabilistic classifier. • Limitation: no pruning or weighting of different tables.Can use logistic regression to learn weights (Bina, Schulte et al. 2013). Bina, B.; Schulte, O.; Crawford, B.; Qian, Z. & Xiong, Y. “Simple decision forests for multi-relational classification”, Decision Support Systems, 2013

  18. Thank you! • Any questions? A Hierarchy of Independence Assumptions

More Related