{"id":730,"date":"2017-03-12T12:11:00","date_gmt":"2017-03-12T12:11:00","guid":{"rendered":"http:\/\/www.kelp-ml.org\/?page_id=730"},"modified":"2017-03-13T17:39:25","modified_gmt":"2017-03-13T17:39:25","slug":"kernels-on-pairs","status":"publish","type":"page","link":"http:\/\/www.kelp-ml.org\/?page_id=730","title":{"rendered":"Kernels on Pairs"},"content":{"rendered":"

They are implementations of the class KernelOnPairs<\/a>\u00a0and operate on instances of the class ExamplePair<\/a>.<\/p>\n

\n

Preference Kernel<\/h3>\n

Java class<\/strong>: PreferenceKernel<\/p>\n

Source code: PreferenceKernel.java<\/a><\/p>\n

Maven Project<\/strong>: kelp-core<\/a><\/p>\n

JSON type<\/strong>:\u00a0preference<\/p>\n

Description<\/strong>: in the learning to rank scenario, the preference kernel (Shen and Joshi, 2003) compares two pairs of ordered objects \"p_a = \langle a_1, a_2 \rangle\" and \"p_b = \langle b_1, b_2 \rangle\":<\/p>\n

\"K ( p_a, p_b ) = BK(a_1, b_1) + BK(a_2, b_2) - BK(a_1, b_2) - BK(a_2, b_1)\"<\/p>\n

where BK<\/em> is a generic kernel operating on the elements of the pairs. The underlying idea is to evaluate whether the first pair\"\langle a_1, a_2 \rangle\" aligns better to the second pair in its regular order\"\langle b_1, b_2 \rangle\" rather than to its inverted order\"\langle b_2, b_1 \rangle\".<\/p>\n

Parameters<\/strong>:<\/p>\n