Using Text Mining to Infer Semantic Attributes for Retail Data Mining

1
Using Text Mining to Infer Semantic Attributes
for Retail Data Mining

• Authors Rayid Ghani Andrew E. Fano
• Presenter Vishal Mahajan
• INFS795

2
Agenda

• Drawbacks in Current Data Mining Techniques.
• Purpose.
• Assumptions and Constraints.
• Methodology or Approach.
• Extraction of Feature Set.
• Labeling .
• Classification Techniques.
• Naïve Bayes
• EM
• Experimental Results.
• Recommender System.

3
Drawbacks in Current Data Mining Techniques

• Semantic Features not automatically considered.
• Transactional Data analyzed without analyzing the
  customer.
• Trending is partial.
• Retail Items treated as objects with no
  associated semantics.
• Data Mining Techniques (association rules,
  decision trees, neural networks) ignore the
  meaning of items and semantics associated with
  them.

4
Purpose of the Presentation

• Describe a system that extracts semantic
  features.
• Populate the knowledge base with the semantic
  features.
• Use of text mining in retailing to extract
  semantic features from website of retailers.
• How profiles of customers or group of customers
  can be build using Text Mining.

5
Assumptions Constraints

• Focus on Apparel Retail segment only.
• Results focus on extracting those semantic
  features that are deemed important by CRM or
  Retail experts.
• Data extracted from retailers website.
• Models generated can be extended beyond the
  Apparel Retail segment.

6
Approach

• Collect Information about products.
• Define set of features to be extracted.
• Label the data with values of the features.
• Train a classifier/extractor to use the labeled
  training to extract features from unseen data.
• Extract Semantic Features from new products by
  using trained classifier.
• Populate a knowledge base with the products and
  corresponding feature.

7
Data Collection Methodology

• Use of web crawler to extract the following from
  large retailers website
• Names
• URLs
• Description
• Prices
• Categories of all Products Available
• Use of wrappers.
• Extracted Information stored in a database and a
  subset chosen.

8
Extraction of Feature Set

• Feature selection based on Expert Systems.
• Use of extensive domain knowledge.
• Feature selection based on Retail Apparel section
  in mind.
• Feature Selected for the project ?
• Age Group
• Functionality
• Price
• Formality
• Degree of Conservativeness
• Degree of Sportiness
• Degree of Trendiness
• Degree of Brand Appeal

9
Labeling Training Data

• Database created with data from collected from
  retailer website.
• Subset of 600 products chosen and labeled.
• Labeling guidelines provided

10
Details of Features extracted from each Product
Description
11
Verifying Training Data

• Disjoint Dataset as labeling done by different
  individuals.
• Association rules (between features) used to
  obtain consistency in labeled data.
• Apriori algorithm
• Apriori Algorithm implemented with single and two
  feature antecedents and consequents.
• Desired Consistency in Labeling achieved by
  applying associating rules

12
Apriori Algorithm

• Find the frequent itemsets the sets of items
  that have minimum support
• A subset of a frequent itemset must also be a
  frequent itemset
• i.e., if AB is a frequent itemset, both A and
  B should be a frequent itemset
• Use the frequent itemsets to generate association
  rules.

13
The Apriori Algorithm Example
L1
C1
Scan D
C2
Database D
C2
L2
Scan D
L3
C3
Scan D
14
Training from Labeled Data

• Learning problem treated as a text classification
  problem.
• Only one text classifier for each semantic
  feature.
• e.g Price of product will be classified as either
  discount or average or luxury.
• Age group is classified as Juniors or Teens or
  GenX or Mature or All Ages.
• Classification was performed using Naïve Bayes
  classification.

15
Sample Association Rules
16
Naïve Bayes

• Simple but effective text classification method.
• Class is selected according to class prior
  probabilities.
• This Model assumes each word in a document is
  generated independently of the other in the
  class.

where N(wt,di) count of times word wt occurs in
document di and Pr(cj,di) 0,1)
17
Incorporating Unlabeled Data

• Initial sample was for 600 products only.
• Need to take care of unlabeled products to make
  any meaningful predictions.
• Use of Supervised learning algorithms.
• These algorithms have proved to reduce the
  classification error considerably.
• Use of Expectation-Maximization (EM) Algorithm as
  the supervised technique.

18
Expectation-Maximization (EM) Method

• EM is an iterative statistical technique for
  maximum likelihood estimation for incomplete
  data.
• In the retail classification problem, unlabeled
  data is considered as incomplete data.
• EM ?
• Locally maximizes the likelihood of the
  parameter.
• Gives estimates for missing values.

19
Expectation-Maximization (EM) Method- cont

• EM method is a 2-step process.
• Initial Parameters are set using naïve Bayes from
  just the labeled documents.
• Subsequent iteration of E- and M-Steps.
• E-Step
• Calculates probabilistically weighed class label
  Pr(cjdj), for every unlabeled document.
• M-Step
• Estimates new classifier parameter using all
  documents (Equation 1).
• E and M steps iterated unless classifier
  converges

20
Experimental Results
21
Experimental Results
22
Results on new data set

• The subset of data that was used earlier was from
  a single retailer.
• Another sample of data was collected from variety
  of retailers. The results are as follows.
• Results are consistently better.

23
Recommender System

• Creation of customer profiles (real time) is
  feasible by analyzing the text associated with
  products and by mapping it to pre-defined
  semantic features.
• Identity of customer is not known and prior
  transaction history is unknown.
• Semantic features are inferred by the browsing
  pattern of the customer.
• Helps in suggesting new products to the customers.

24
Recommender System

• Mathematically ?
• P(AijProduct)
• Where Aij is the jth value of ith attribute
• isemantic attributes, jpossible values
• User profile is constructed as follows
• Pr(Ui,jPast N Items) 1/N

i,j
is calculated
25
Types of Recommender Systems

• Two Types of Recommender Systems.
• Collaborative Filtering.
• Collect user feedback in terms of ratings.
• Exploit similarities and differences of customers
  to recommend items.
• Issues
• Sparsity Problem.
• New Items.
• Content Filtering
• Compares the contents
• Issues
• Narrow in scope
• Recommends similar products only

26
Conclusions

• The systems learns from the use of supervised and
  semi-supervised techniques.
• Major assumptions..Products accurately convey the
  semantic attributes.??
• Small sample of data used to Infer results.
  Practical applications not verified.
• System bootstrapped from a small number of
  labeled training examples.
• Interesting application which could be evolved to
  generate trends for retail marketers.