PhD Defense -- Ashish Mangalampalli
Download as PPTX, PDF•1 like•912 views
The document presents a fuzzy associative rule-based approach for pattern mining and classification, outlining techniques such as fuzzy pre-processing, associative classification methods (like SEAC and FSEAC), and their applications in object classification and advertising. It emphasizes the benefits of using fuzzy logic to enhance classification accuracy by addressing the limitations of crisp classifiers, notably in real-world datasets. The study also includes experimental results showcasing the performance of these techniques in various domains.
![Crisp Associative Classification
Most associative classifiers are crisp
Most real-life datasets contain binary and numerical attributes
Use sharp partitioning
Transform numerical attributes to binary ones, e.g. Income =
[100K and above]
Drawbacks of sharp partitioning
Introduces uncertainty, especially at partition boundaries
Small changes in intervals lead to misleading results
Gives rise to polysemy and synonymy
Intervals do not generally have clear semantics associated
For example, sharp partitions for the attribute Income
Up to 20K, 20K-100K, 100K and above
Income = 50K would fit in the second partition
But, so would Income = 99K
5](https://image.slidesharecdn.com/ashish-phddefense-120728174742-phpapp01/85/PhD-Defense-Ashish-Mangalampalli-5-320.jpg)
![Fuzzy Associative Classification
Fuzzy logic
Used to convert numerical attributes to fuzzy attributes
(e.g. Income = High)
Maintains integrity of information conveyed by numerical
attributes
Attribute values belong to partitions with some
membership - interval [0, 1]
6](https://image.slidesharecdn.com/ashish-phddefense-120728174742-phpapp01/85/PhD-Defense-Ashish-Mangalampalli-6-320.jpg)
PhD Defense -- Ashish Mangalampalli
- 1. A Fuzzy Associative Rule- based Approach for Pattern Mining and Pattern-based Classification Ashish Mangalampalli Advisor: Dr. Vikram Pudi Centre for Data Engineering International Institute of Information Technology (IIIT) Hyderabad 1
- 2. Outline Introduction Crisp and Fuzzy Associative Classification Pre-Processing and Mining Fuzzy Pre-Processing – FPrep Fuzzy ARM – FAR-Miner and FAR-HD Associative Classification – Our Approach FACISME – Fuzzy Adaption of ACME (Maximum Entropy Associative Classifier) Simple and Effective Associative Classifier (SEAC) Fuzzy Simple and Effective Associative Classifier (FSEAC) Associative Classification – Applications Efficient Fuzzy Associative Classifier for Object Classes in Images (I-FAC) Associative Classifier for Ad-targeting Conclusions 2
- 3. Introduction Associative classification Mines huge amounts of data Integrates Association Rule Mining (ARM) with Classification A = a, B = b, C = c → X = x Associative classifiers have several advantages Frequent itemsets capture dominant relationships between items/features Statistically significant associations make classification framework robust Low-frequency patterns (noise) are eliminated during ARM Rules are very transparent and easily understood Unlike black-box-like approach used in popular classifiers, such as SVMs and Artificial Neural Networks 3
- 4. Outline Introduction Crisp and Fuzzy Associative Classification Pre-Processing and Mining Fuzzy Pre-Processing – FPrep Fuzzy ARM – FAR-Miner and FAR-HD Associative Classification – Our Approach Simple and Effective Associative Classifier (SEAC) Fuzzy Simple and Effective Associative Classifier (FSEAC) Associative Classification – Applications Efficient Fuzzy Associative Classifier for Object Classes in Images (I-FAC) Associative Classifier for Ad-targeting Conclusions 4
- 5. Crisp Associative Classification Most associative classifiers are crisp Most real-life datasets contain binary and numerical attributes Use sharp partitioning Transform numerical attributes to binary ones, e.g. Income = [100K and above] Drawbacks of sharp partitioning Introduces uncertainty, especially at partition boundaries Small changes in intervals lead to misleading results Gives rise to polysemy and synonymy Intervals do not generally have clear semantics associated For example, sharp partitions for the attribute Income Up to 20K, 20K-100K, 100K and above Income = 50K would fit in the second partition But, so would Income = 99K 5
- 6. Fuzzy Associative Classification Fuzzy logic Used to convert numerical attributes to fuzzy attributes (e.g. Income = High) Maintains integrity of information conveyed by numerical attributes Attribute values belong to partitions with some membership - interval [0, 1] 6
- 7. Outline Introduction Crisp and Fuzzy Associative Classification Pre-Processing and Mining Fuzzy Pre-Processing – FPrep Fuzzy ARM – FAR-Miner and FAR-HD Associative Classification – Our Approach Simple and Effective Associative Classifier (SEAC) Fuzzy Simple and Effective Associative Classifier (FSEAC) Associative Classification – Applications Efficient Fuzzy Associative Classifier for Object Classes in Images (I-FAC) Associative Classifier for Ad-targeting Conclusions 7
- 8. Pre-Processing and Mining Fuzzy pre-processing Convert crisp dataset (binary and numerical attributes) into fuzzy dataset (binary and fuzzy attributes) FPrep Algorithm used Efficient and robust Fuzzy ARM algorithms Web-scale datasets mandate such algorithms Fuzzy Apriori is most popular Many efficient crisp ARM algorithms exist like ARMOR and FP-Growth Algorithms used FAR-Miner for normal transactional datasets FAR-HD for high dimensional datasets 8
- 9. Outline Introduction Crisp and Fuzzy Associative Classification Pre-Processing and Mining Fuzzy Pre-Processing – FPrep Fuzzy ARM – FAR-Miner and FAR-HD Associative Classification – Our Approach Simple and Effective Associative Classifier (SEAC) Fuzzy Simple and Effective Associative Classifier (FSEAC) Associative Classification – Applications Efficient Fuzzy Associative Classifier for Object Classes in Images (I-FAC) Associative Classifier for Ad-targeting Conclusions 13
- 10. Associative Classification – Our Approach AC algorithms like CPAR and CMAR only mine frequent itemsets Processed using additional (greedy) algorithms like FOIL and PRM Overhead in running time; process more complex Association rules directly used for training and scoring Exhaustive approach Controlled by appropriate support Not a time-intensive process Rule pruning and ranking take care of huge volume and redundancy Classifier built in a two-phased manner Global rule-mining and training Local rule-mining and training Provides better accuracy and representation/coverage 14
- 11. Associative Classification – Our Approach (cont’d) Pre-processing to generate fuzzy dataset (for fuzzy associative classifiers) using FPrep Classification Association Rules (CARs) mining using FAR-Miner or FAR-HD CARs pruning and classifier training using SEAC or FSEAC Rule ranking and application (scoring) techniques 15
- 12. Simple and Effective Associative Classifier (SEAC) Direct mining of CARs – faster and simpler training CARs used directly through effective pruning and sorting Pruning and rule-ranking based on Information gain Rule-length Two-phased manner Global rule-mining and training Local rule-mining and training 16
- 13. SEAC - Example Example Dataset Scoring Example Unlabeled: B=2, C=2 X=1 → 16, 17, 19 (IG=0.534) X=2 → 13, 14, 20 (IG=0.657) Ruleset 17
- 14. Fuzzy Simple and Effective Associative Classifier (FSEAC) Amalgamates Fuzzy Logic with Associative Classification Pre-processed using FPreP CARs mined using FAR-Miner / FAR-HD CARs pruned based on Fuzzy Information Gain (FIG) and rule length - no sorting required Scoring – rules applied taking µ into account Sorting done then Final score computed 18
- 15. FSEAC - Example Format for Fuzzy Version of Dataset Example Dataset Fuzzy Version of Example Dataset 19
- 16. FSEAC – Example (cont’d) Ruleset 20
- 17. SEAC and FSEAC Experimental Setup SEAC 12 classifiers (Associative and non-associative) 14 UCI ML datasets 100-5000 records per dataset 2-10 classes per dataset Up to 20 features per dataset 10-fold Cross Validation FSEAC 17 classifiers (Associative and non-associative; fuzzy and crisp) 23 UCI ML datasets 100-5000 records per dataset 2-10 classes per dataset Up to 60 features per dataset 10-fold Cross Validation 21
- 18. SEAC – Results (10 fold-CV) continued 22
- 19. SEAC - Results (10 fold-CV) 23
- 20. FSEAC - Results (10 fold-CV) continued 24
- 21. FSEAC - Results (10 fold-CV) 25
- 22. Outline Introduction Crisp and Fuzzy Associative Classification Pre-Processing and Mining Fuzzy Pre-Processing – FPrep Fuzzy ARM – FAR-Miner and FAR-HD Associative Classification – Our Approach Simple and Effective Associative Classifier (SEAC) Fuzzy Simple and Effective Associative Classifier (FSEAC) Associative Classification – Applications Efficient Fuzzy Associative Classifier for Object Classes in Images (I-FAC) Associative Classifier for Ad-targeting Conclusions 26
- 23. Efficient Fuzzy Associative Classifier for Object Classes in Images (I-FAC) Adapts fuzzy associative classification for Object Class Detection in images Speeded-Up Robust Features (SURF) - interest point detector and descriptor for images Fuzzy clusters used as opposed to hard clustering used in Bag- of-words Only positive class (CP) examples used for mining Negative class (CN) in object class detection is very vague CN = U – CP Rules are pruned and ranked based on Information Gain Other AC algorithms use third-party algorithms for rule- generation from frequent itemsets Top k rules are used for scoring and classification 27 ICPR 2010
- 24. I-FAC SURF points extracted from positive class images FCM applied to derive clusters Clusters (with µs) used to generate dataset for mining 100 fuzzy clusters as opposed to1000-2000 crisp clusters-based algorithms ARM generates Classification Association Rules (CARs) associated with positive class CARs are pruned and sorted using Fuzzy Information Gain (FIG) of each rule Length of each rule i.e. number of attributes in each rule Scoring based on rule-match and FIG 28 ICPR 2010
- 25. I-FAC - Performance Study Performs well when compared to BOW or SVM Very well at low FPRs (≤0.3) Fuzzy nature helps avoid polysemy and synonymy Uses only positive class for training 30 ICPR 2010
- 26. Visual Concept Detection on MIR Flickr Revamped version of I-FAC Multi-class detection 38 visual concepts e.g. car, sky, clouds, water, building, sea, face Experimental evaluation First 10K images of MIR Flick dataset AUC values for each concept 31
- 27. Experimental Results (3-fold CV) continued 32
- 28. Experimental Results (3-fold CV) 33
- 29. Look-alike Modeling using Feature-Pair- based Associative Classification Display-ad targeting currently done using methods which rely on publisher-defined segments like Behavior-targeting (BT) Look-alike model trained to identify similar users Similarity is based on historical user behavior Model iteratively rebuilt as more users are added Advertiser supplies seed list of users Approach for building advertiser specific audience segments Complements publisher defined segments such as BT Provides advertisers control over the audience definition Given a list of target users (e.g., people who clicked or converted on a particular category or ad campaign), find other similar users. 34 WWW 2011
- 30. Look-alike Modeling using Feature-Pair- based Associative Classification – cont’d Enumerate all feature-pairs in training set occurring in at least 5 positive-class records Feature-pairs modelled as AC rules Only rules for positive class used Works well in Tail Campaigns Affinity measured by Frequency-weighted LLR (F-LLR) FLLR = P(f) log(P(f | conv) / P(f | non-conv)) Rules sorted in descending order by F-LLRs Scoring - Top k rules are applied Cumulative score from all rules used for classification 35 WWW 2011
- 31. Performance Study Two pilot campaigns 300K records each Lift Baseline (Conversion Lift (AUC) One record per user Rate) Training window - 14 Random days 82% – Targeting Scoring window - seven Linear SVM 301% 11% days GBDT 100% 2% Works very well for Tail Results on a Tail Campaign Campaigns Can find meaningful Lift Baseline Lift (Conversion Rate) associations in extremely (AUC) sparse and skewed data Random 48% – Targeting SVM and GBDT work Linear SVM -12% -6% well for Head Campaigns GBDT -40% -14% Results on a Head Campaign 36 WWW 2011
- 32. Outline Introduction Crisp and Fuzzy Associative Classification Pre-Processing and Mining Fuzzy Pre-Processing – FPrep Fuzzy ARM – FAR-Miner and FAR-HD Associative Classification – Our Approach Simple and Effective Associative Classifier (SEAC) Fuzzy Simple and Effective Associative Classifier (FSEAC) Associative Classification – Applications Efficient Fuzzy Associative Classifier for Object Classes in Images (I-FAC) Associative Classifier for Ad-targeting Conclusions 37
- 33. Conclusions Fuzzy pre-processing for dataset transformation Fuzzy ARM for various types of datasets Fuzzy and Crisp Associative Classifiers for various domains Customizations required for different domains Pre-processing Pruning Rule ranking techniques Rule application (scoring) techniques 38
- 34. References Ashish Mangalampalli, Adwait Ratnaparkhi, Andrew O. Hatch, Abraham Bagherjeiran, Rajesh Parekh, and Vikram Pudi. A Feature-Pair-based Associative Classification Approach to Look-alike Modeling for Conversion-Oriented User-Targeting in Tail Campaigns. In International World Wide Web Conference (WWW), 2011. Ashish Mangalampalli, Vineet Chaoji, and Subhajit Sanyal. I-FAC: Efficient fuzzy associative classifier for object classes in images. In International Conference on Pattern Recognition (ICPR), 2010. Ashish Mangalampalli and Vikram Pudi. FPrep: Fuzzy clustering driven efficient automated pre-processing for fuzzy association rule mining. In IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), 2010. Ashish Mangalampalli and Vikram Pudi. FACISME: Fuzzy associative classification using iterative scaling and maximum entropy. In IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), 2010. Ashish Mangalampalli and Vikram Pudi. Fuzzy Association Rule Mining Algorithm for Fast and Efficient Performance on Very Large Datasets. In IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), 2009. 39
- 35. Thank You, and Questions 40