Identifying customers who are likely to respond to a product offering is an important issue in direct marketing.Response models are typically built from historical purchase data. A popular method of choice, logistic regression, is easy to understand and build, but limited in that the model is linear in parameters. Neural networks are nonlinear and have been found to improve predictive accuracies for a variety of business applications. Neural networks have not always demonstrated clear supremacy over traditional statistics competitors, largely because of over-fitting and instability. Combining multiple networksmay alleviate these problems. A systematic method of combining neural networks is proposed, namely bagging or bootstrap aggregating, whereby overfitted multiple neural networks are trained with bootstrap replicas of the original data set and then averaged. We built response models using a publicly available DMEF data set with three methods: bagging neural networks, single neural networks, and conventional logistic regression. The proposed method not only improved but also stabilized the prediction accuracies over the other two.

This paper focuses on performance evaluation of wireless networked control system (NCS) via IEEE 802.15.4 with time-triggered contention-based carrier sensing multiple access with collision (CSMA/CA) algorithm to satisfy the industrial quality-of-service (QoS). Although the IEEE 802.15.4 is considered to a promising protocol for wireless NCS, its performance is not stable and its delay is not bounded under heavy traffic because its medium access control (MAC) is based on the contention-based CSMA/CA scheme. As an alternative approach to avoid these limitations, this paper presents a time-triggered CSMA/CA algorithm for IEEE 802.15.4 non-beacon mode that relies on a time division multiplexing access (TDMA) method. The time-triggered CSMA/CA algorithm for IEEE 802.15.4 non-beacon mode was implemented an experimental wireless NCS, and its control performance was compared with conventional CSMA/CA algorithm. Based on the experimental results, the time-triggered CSMA/CA algorithm is a feasible alternative for wireless NCS.

Traffic accidents are still increasing even though vehicles are becoming more intelligent to enhance driver convenience and safety. Single car-on-car rear impacts in urban areas have increased rapidly due to driver inattention. According to a Road Traffic Authority (ROTA) report in Korea in 2006, 85.2% of single car-on-car rear impact accidents occurred at less than 60 km/h, and 25.3% of the total occurred at between 30 km/h and 50 km/h. To prevent rear vehicle crashes in urban areas, automobile manufacturers have developed various low-speed, close-range collision-warning systems. This paper presents a low-speed, close-range collision-warning algorithm for urban areas using fuzzy inference. Experiments using an embedded microprocessor in the driving track demonstrated the feasibility of the proposed collision-warning system. The results indicate that the fuzzy inference-based, low-speed, close-range collision-warning system could reduce traffic accidents by alerting the driver to potential collisions.

No abstract