A Backpropagation Neural Network Model on Precipitation Forecasting in the Philippines
Backpropagation neural networks were used to forecast daily rainfall with minimal error for Metro Manila in order to have an inexpensive way of accurately predicting weather. Calamities brought on by heavy rainfall have caused great economic, infrastructure and human loses. Neural networks have the ability to discern complex patterns in noisy data; this makes it a viable method for weather forecasting. Daily precipitation, humidity, rain indication, sea level pressure, temperature and maximum sustained wind speed for January 2000 to December 2010 were acquired from the Philippine Atmospheric Geophysical and Astrological Services Administration. The neural network made use of Python 2.7.2 and the backpropagation program by Neil Schemenauer (python.org). It considered different neural network architectures with a total of 2844 data sets for training and 708 data sets for testing. Each neural network’s accuracy was measured with a graph of the actual and predicted values, correlation coefficient, and root mean square error. It was observed that the neural network with architecture 5-8-1 yielded the most accurate results as it had the highest correlation coefficient of 0.48599 and smallest root mean square error of 14.84. It was also observed that the trends of the predicted values followed that of the target values. This suggests that it is possible to create a neural network with a moderate correlation given daily weather data. It is recommended that further researches make use of hourly data instead of daily data for more accurate results. Other variables, which might affect rainfall, not in this study should also be considered. This research could aid in the anticipation of calamities and the decision making involved in shipping, fishing and aviation industries.
Fig Preservation
Figs have become an expanding industry here in New Zealand and are a current export fruit which could potentially provide a large amount of profit to both growers and the New Zealand market as a whole. Nicola’s family has about 10 acres of fig trees. They sell the figs locally and as an export. They generally sell for about $13 per kilogram here in New Zealand and $26 in the USA. However, figs only have a shelf life of about 7 days. This is because at present there is no proven pre or post-harvest treatment or method of storage that helps to decrease the rate of decay of the fig fruit. After researching post-harvest treatments for figs, Nicola found a report which claimed to have developed treatments that increased the shelf life of figs by about 5 weeks. With this kind of increase, it would be possible to transport, store and export figs over longer periods of time without running the risk of losing large amounts of produce, or delivering unsatisfactory fruit to customers. Nicola developed 7 different post-harvest treatments based on the ones that had shown promise in earlier research. These were hot-water baths of different temperatures, both with and without different bleach concentrations. To test these on the fruit she set up four experiments – a dry matter test, a firmness test (using a penetrometer), a colour test and observation of detrimental features of the fig. She tested these treatments at 0, 7, 14, 21 and 28 days from harvest. Nicola found that after 7 days, the firmness of all of the figs that had been treated had decreased to a large degree. The only figs that did not have a massive decrease were the untreated fruit. However after about 14 days, the firmness of all of the fruit became about the same and after this 14 day mark, she would not have considered any of the figs to be edible. However, in the appearance tests, it seemed that the treated figs that had the least amount of mould and rot were the ones that had been treated with higher levels of bleach such as the 55 degree Celsius water bath with 0.003L of bleach to every litre of water, and the 35 degree Celsius water bath with the same concentration of bleach. Overall, Nicola’s results showed that the hot water bath, and hot water bath and bleach post-harvest treatments did not slow the decay of the fruit in the earlier weeks after picking. In effect, Nicola’s research showed that the information she had relied on to help plan her study had claimed too much and that the treatments were less effective than had been stated. More research will be needed to find a more reliable way to improve the shelf life of figs.
阿里巴巴轉盤問題
本作品為環球數學城市競賽的考題之推廣（ International Mathematics Tournament of the Towns, Senior A-Level Paper, Fall 2009, No. 7 ），然而此題目本身比較接近Scientific American （ Feb 1979 ）中 Martin Gardner 的文章 The Rotating Table 所提出的問題。 而此問題的多邊形版已被數學家解決(Ted Lewis & Stephen Willard, 1980, The Rotating Table,Mathematics Magazine, 53,Page 174-175.)。然而此份作品採用簡潔的初等數學歸納法證明了原本的多邊形情形，且專注於探討條件改變為m=n-1時的多邊形情形。 m=n-1時的情形是前人所沒有做出結果的，且m=n-1的情形遠遠比m=n 的情況來得複雜多變，其中估計上下界所使用的許多組合技巧相當特別。然而其中當n為3 和5的倍數時有比較特殊的情況，這一個部分僅能給出目前得出的k值上界。而其他情況則能夠找出所有的k值。