On the Parzen Kernel-Based Probability Density Function Learning Procedures Over Time-Varying Streaming Data With Applications to Pattern Classification.

In this paper, we propose a recursive variant of the Parzen kernel density estimator (KDE) to track changes of dynamic density over data streams in a nonstationary environment. In stationary environments, well-established traditional KDE techniques have nice asymptotic properties. Their existing extensions to deal with stream data are mostly based on various heuristic concepts (losing convergence properties). In this paper, we study recursive KDEs, called recursive concept drift tracking KDEs, and prove their weak (in probability) and strong (with probability one) convergence, resulting in perfect tracking properties as the sample size approaches infinity. In three theorems and subsequent examples, we show how to choose the bandwidth and learning rate of a recursive KDE in order to ensure weak and strong convergence. The simulation results illustrate the effectiveness of our algorithm both for density estimation and classification over time-varying stream data.

Seasonal variation in the amount of dietary carbohydrate not absorbed from the intestine after breakfast in elderly Japanese females.

It was previously shown that there is seasonality in the amount of dietary carbohydrate not absorbed from the intestine after breakfast, the amount of carbohydrate in winter being significantly larger than that in autumn in young Japanese subjects. In order to investigate this phenomenon further, the experiment was repeated on 22 elderly Japanese female subjects (61-78 yrs of age) during the four seasons of the year. The amount of unabsorbed dietary carbohydrate by the breath hydrogen test, which measures the amount of hydrogen in exhaled air, was then estimated. A 6 g solution of lactosucrose, an indigestible trisaccharide, was used for comparison. Two groups of subjects, 16 subjects in Osaka and 6 subjects in Nagano, were studied in the summer (July to August) and autumn (October to November) of 2005 and the winter (January to February) and spring (April to May) of 2006. The following results were found using the pooled data of the total of 22 subjects. With regard to the amount of breath hydrogen excretion of the lactosucrose solution, there was no significant difference between the four seasons. There was a significant seasonal change in the efficiency of dietary carbohydrate absorption from the intestine after breakfast. The percentage of total carbohydrate that was not absorbed was lowest in the spring and highest in the winter. A comparison of the results from studies on the elderly and young subjects revealed the percentage of total carbohydrate that was not absorbed in the elderly was significantly lower than in the young in the winter, spring, and summer. These results indicate that there is seasonal variation in the efficiency of dietary carbohydrate absorption from the intestine among elderly female Japanese subjects as well as young female Japanese subjects. They also suggest that the efficiency of dietary carbohydrate absorption from the intestine after breakfast is retained in these naturally active and healthy elderly subjects.

Comparison of carbohydrate digestion between Japanese and Polish healthy subjects.

We have revealed that light environment affects digestion and absorption of dietary carbohydrates in the gastrointestinal tract. This experimental result supposes that the efficiency of carbohydrate absorption may differ among people who live in different latitudes, such as Japanese and Polish people, at the same calendar season. In order to prove this hypothesis, we have been comparing the efficiency of carbohydrate absorption using the breath hydrogen test in Japan and Poland. Here, we report the comparison of the result obtained in the summer of 2004 as the following; (1) Orocecal transit time (OCTT) for indigestible trisaccharide of Japanese subjects was significantly longer than that in Poland (p = 0.043). (2) On the ingestion of minestrone, the amount of unabsorbed carbohydrate of Japanese subjects (which was estimated as trisaccharide equivalent) was significantly larger than that of Polish subjects (p = 0.006).

Seasonal variation in amount of unabsorbed dietary carbohydrate from the intestine after breakfast in Japanese subjects.

We previously showed that daytime dim-light exposure has a negative effect on the efficiency of dietary carbohydrate absorption in the evening, whereas evening-time dim-light exposure has a beneficial effect. These results suggest that seasonal changes in the environmental light may affect gastrointestinal activity, and that there might, therefore, be seasonal variation in the efficiency of dietary carbohydrate absorption from the intestine. In order to prove this hypothesis, we measured the amount of dietary carbohydrate unabsorbed from the intestine after a breakfast in healthy female Japanese subjects during the four seasons of the year. We estimated the amount of unabsorbed dietary carbohydrate by the breath hydrogen test, which measures the amount of hydrogen in exhaled air. A 6 g solution of lactosucrose, an indigestible trisaccharide, was used for comparison. Two groups of subjects, 12 subjects in Osaka and 14 subjects in Nagano, were studied in the winter (January to February), spring (April to May), summer (July to August), and autumn (October to November) of 2004. We found the following results: (1) In no season were there any significant differences between the two subgroups in the orocecal transit time of the breakfast and the lactosucrose solution. Nor were there any significant differences in the amount of unabsorbed dietary carbohydrate from the breakfast. (2) Using the pooled data of the total of 26 subjects, there was no significant seasonal variation in the orocecal transit time of the breakfast or the lactosucrose solution. (3) There was a significant seasonal variation in the amount of unabsorbed dietary carbohydrate from the breakfast. (4) The amount of unabsorbed dietary carbohydrate from the breakfast was largest in winter and smallest in autumn. Results in spring and in summer were similar and intermediate between those in winter and autumn. Post hoc multiple comparison tests showed that the amount of unabsorbed dietary carbohydrate in winter was significantly larger than in autumn. (5) In winter, the average ratio of the amount of unabsorbed dietary carbohydrate to the total amount of carbohydrate in the breakfast was about 12%; in autumn it was about 6%. These results clearly show that there is seasonal variation in the efficiency of intestinal dietary carbohydrate absorption among young female Japanese subjects.