Argument Based Machine Learning


Argument Based Machine Learning (ABML) is a novel approach to machine learning, where classical machine learning is extended with concepts from the field of argumentation. Usually the problem of learning from examples is stated as:

To illustrate the idea of argument based machine learning, consider a simple problem: learning about credit approval. Each example is a customer's credit application together with the manager's decision about credit approval. Each customer has a name and three attributes: PaysRegularly (with possible values yes and no), Rich (possible values yes and no) and HairColor (black, blond, ...). The class is CreditApproved (with possible values yes and no). Let there be three learning examples as shown in the table below.

Credit approval learning examples
Miss Whiteyesnoblondyes

A typical rule learning algorithm will induce the following rule from this data:

IF HairColor = blond THEN CreditApproved = yes

This rule looks good because it is short and it correctly covers two of the three given examples. On the other hand, the rule may not make much sense to a financial expert. We will now look at how this may change when arguments are introduced.

With arguments, the learning problem statement changes to:

To illustrate what it means to "explain the examples using given arguments", consider again the data from Table and assume that an expert gave an argument for Mrs. Brown:

Mrs. Brown received credit because she is rich.

Now consider again the rule above that all blond people receive credit. This rule correctly classifies Mrs. Brown, but it does not explain this classification in terms of the given argument for Mrs. Brown. The rule does not even mention Mrs. Brown's property in the argument, namely that she is rich. Therefore we say that this rule is not consistent with arguments given, and an ABML algorithm has to induce another rule to this effect. Therefore an argument based rule learning algorithm would induce:

IF Rich=yes THEN CreditApproved=yes,

which is clearly consistent with argument that Mrs. Brown received credit because she is rich.

The motivation for ABML bases on the fact that the fundamental problem of machine learning is dealing with large spaces of possible hypotheses, which is traditionally solved by either biasing learning towards simpler hypotheses, i.e. applying Occam's razor, or by using experts' domain knowledge for constraining search. ABML is special algorithm for using expert's domain knowledge, where experts articulate their knowledge with respect to a chosen learning example. Standard machine learning methods can only accept general prior knowledge (like ILP), however it has been shown that experts face difficulties when providing general knowledge about theory underlying learning examples. Thus, it is much easier to acquire arguments from experts than general rules. Moreover, the ABML algorithm has been shown to improve on the performance of existing machine learning algorithms on several data sets.

Further reading about basics of argument based machine learning can be found in [1,2]. Papers [3] and [4] demonstrate ABML on two real-world use cases. Paper about extreme value corrections (EVC) [5] is not directly related to argument based learning, however it is crucial for evaluation quality of rules learned from arguments. For more about EVC see link.


To be added.


[1] Martin Mozina, Jure Zabkar, Ivan Bratko. Argument Based Machine Learning. AI Journal. Download

[2] Martin Mozina, Jure Zabkar, Ivan Bratko. Implementation of and experiments with ABML and MLBA. ASPIC deliverable D3.4. Download

[3] Martin Mozina, Jure Zabkar, Trevor Bench-Capon, Ivan Bratko. Argument Based Machine Learning Applied to Law. Artificial Intelligence and Law. Download

[4] Jure Zabkar, Martin Mozina, Ivan Bratko. Argument Based Machine Learning in Medical Domain. COMMA 2006 Download

[5] Martin Mozina, Janez Demsar, Jure Zabkar, Ivan Bratko. Why is Rule Learning Optimistic and How to Correct It. European Conference on Machine Learning (2006). Download

[6] Martin Mozina, Matej Guid, Jana Krivec, Aleksander Sadikov, Ivan Bratko. Fighting knowledge acquisition bottleneck with argument based machine learning. ECAI (2008). Download

[7] Martin Mozina, Matej Guid, Aleksander Sadikov, Vida Groznik, Jana Krivec, Ivan Bratko. Conceptualizing Procedural Knowledge Targeted at Students with Different Skill Levels. Unpublished. Download