_PROBLEM CoEPrA-2006_Regression_001 _GROUP_NAME Scott Oloff _GROUP_MEMBERS Scott Oloff _ADDRESS Boehringer Ingelheim 900 Ridgebury Rd. Ridgefield, CT 06877 _MODELING_PROCEDURE I used a newly developed machine learning approach called kScore, which is based on kNN and SVM technologies. The approach is kernel free and uses all training compounds in a distance-weighted manner to predict a test compound rather than just its k Nearest Neighbors. The algorithm also incorporates Generalization constraints, a soft margin classifier (for Classification), and an Epsilon Loss function (for Regression) similar to those proposed by V. Vapnik. There is a built in Applicability Domain for each kScore model and a more detailed explanation of the method is being submitted for publication. The descriptors of the original training set were normalized from 0 to 1. The training set was then randomly divided into 5 training and test sets whereby the test set contained 20% of the compounds. For each training/test split several kScore models were generated (using the COEPRA provided descriptors) to find the best values of C and Epsilon for external prediction. The most predictive models were collected and used in a consensus fashion to predict the external test set. The average LOO Training Q2's were 0.78-0.85 and the average Test R2's between 0.78 and 0.88. The most predictive values of Epsilon and C varied from 0.1 to 0.4 and 500 to 1000, respectively. The descriptor weights of the most predictive model are attached. The objects that fell just outside of the models' applicability domain but were predicted anyway for the purpose of this competition are: Obj_00008 Obj_00019 Obj_00022 Obj_00045 Obj_00047 Obj_00050 Obj_00057 Obj_00058 Obj_00060 Obj_00079 Obj_00083 Obj_00087 Obj_00088 _PREDICTION Obj_00001 5.577 Obj_00002 5.666 Obj_00003 4.757 Obj_00004 4.694 Obj_00005 7.766 Obj_00006 6.558 Obj_00007 4.944 Obj_00008 5.029 Obj_00009 5.523 Obj_00010 5.529 Obj_00011 6.127 Obj_00012 4.880 Obj_00013 5.656 Obj_00014 4.959 Obj_00015 6.732 Obj_00016 5.563 Obj_00017 4.986 Obj_00018 5.884 Obj_00019 5.050 Obj_00020 4.904 Obj_00021 4.793 Obj_00022 5.528 Obj_00023 5.390 Obj_00024 5.764 Obj_00025 4.965 Obj_00026 6.691 Obj_00027 6.106 Obj_00028 5.749 Obj_00029 5.392 Obj_00030 4.991 Obj_00031 5.562 Obj_00032 5.671 Obj_00033 5.676 Obj_00034 6.041 Obj_00035 6.087 Obj_00036 6.285 Obj_00037 4.934 Obj_00038 5.497 Obj_00039 5.693 Obj_00040 4.494 Obj_00041 5.597 Obj_00042 4.954 Obj_00043 6.923 Obj_00044 4.943 Obj_00045 5.877 Obj_00046 5.769 Obj_00047 4.959 Obj_00048 5.766 Obj_00049 5.015 Obj_00050 4.967 Obj_00051 5.856 Obj_00052 5.681 Obj_00053 5.508 Obj_00054 4.812 Obj_00055 4.996 Obj_00056 5.638 Obj_00057 5.313 Obj_00058 4.904 Obj_00059 7.124 Obj_00060 5.030 Obj_00061 5.825 Obj_00062 4.979 Obj_00063 5.673 Obj_00064 4.935 Obj_00065 5.915 Obj_00066 7.096 Obj_00067 6.233 Obj_00068 5.771 Obj_00069 5.680 Obj_00070 4.784 Obj_00071 4.279 Obj_00072 5.499 Obj_00073 4.981 Obj_00074 4.953 Obj_00075 6.052 Obj_00076 4.478 Obj_00077 5.897 Obj_00078 5.878 Obj_00079 5.132 Obj_00080 6.117 Obj_00081 5.768 Obj_00082 5.623 Obj_00083 5.560 Obj_00084 6.125 Obj_00085 6.992 Obj_00086 5.666 Obj_00087 5.471 Obj_00088 5.511