Ten Holstein cows were sprayed with one of two teat-disinfecting solutions, onecontaining iodine, and the other containing mainly fatty and lactic acids but no iodine. The cows were milked twice ...daily. A six month, double switch-back experimental design was employed which consisted of alternately applying iodine, fatty acid, and iodine disin-fectant treatments for two months each to the cows in group A while the group B cows concurrently received simultaneous alternate treatment of fatty acid, iodine, and fatty acid, also for two months each. Somatic cell counts (SCC) of individual cow were measured monthly by Fosomatic 360. Milk NAGase activity (NAG) was determined four times during the experiment using Nissan fluorometer. Both are used as indicators of udder health. The teat skin conditions, clinical signs of mastitis as well as general health status were observed usually bythe milkers. The means±SD of SCCs and NAG were 57. 4±2. 8× 1, 000/ml versus 56. 7±3. 0× 1, 000/ml and 3. 237±1. 711nmol/min/ml versus 3. 174± 1. 837nmol/min/ml for iodine and no io-dine solutions respectively. No significant differences (p>0.01) in both indicators werefound. However, the teat skin conditions of each cow were better after the non-iodineteat dip solution was applied. This suggests that the non-iodine teat dip solution is amore effective pre-milking teat disinfectant that also eliminates possible iodine contamina-tion of the milk.
We consider the problem of recognizing the driving environment of a vehicle by using information obtained from some sensors of the vehicle. Previously, we presented a recognition algorithm based on ...an usual method of pattern matching by use of fuzzy reasoning. Furthermore, this algorithm was extended to meet the demands of nonstandard drivers and changes of vehicle properties. In this algorithm we supposed that an extra source of knowledge (supervisor) for correcting the decision taken by the classifier could be acquired, and inevitably we expected additional sensors. To cover such weakness we present a self-supervised learning algorithm. Here the supervisor is constructed by using a-cut of the membership function of the representative class. Computer simulation on practical uses of the vehicle shows efficiency of the self-supervised learning.
PDF
