China
Africa
Explore chapters and articles related to this topic
Analysis of Ontology-Based Semantic Association Rule Mining
Published in Archana Patel, Narayan C. Debnath, Bharat Bhushan, Semantic Web Technologies, 2023
Analyzing trends from the huge transactional data is challenging and important in business, medical, banking, academics, and so on. Trend analysis helps to take decisions that necessitate the understanding and analysis of complex data. The association rule mining helps to identify the trends by discovering the relations among data. The relations between the biomedical entities are helpful to understand the relationship between the diseases and their corresponding symptoms including their treatment. The trend breakthrough in customer’s loan may help the bank managers to decide whether the loan can be sanctioned or not. In business, the relations among the items have been used in marketing to promote product sales. The association mining algorithms are used to generate all the possible relations among the entities and the interesting relations can be derived using support and confidence metrics. The association rules are expressed in the form of antecedent and consequent which consists of the items list. For example, the rule depicted in Figure 14.1 shows that 5% of antivirus software is bought with the purchase of a computer and printer, and 60% of the customers who purchased antivirus software also bought a computer and printer.
Air cargo forecasting
Published in Peter S. Morrell, Thomas Klein, Moving Boxes by Air, 2018
Peter S. Morrell, Thomas Klein
Trend analysis techniques use time series and attempt to fit a trend line through historical data, whether on an annual or monthly basis. This line is then projected into the future depending on the equation that best describes the historical data. Statistical techniques can be used ranging from simple averages to the more complex exponential smoothing. There is no attempt to understand the causes of traffic trends, and these methods are not reliable beyond five or so years into the future. Apart from these problems they work on the assumption that past trends will continue into the future. This has not been the case in the past especially with the average capacity of flights which increased quite fast in the 1970s and 1980s only to flatten out in the 1990s and increase again significantly in the 2000s due to more cargo-friendly wide-body passenger aircraft. Similarly trend projections would not have taken into account the rapid growth in low-cost airlines over the past two decades in Europe.
Elements
Published in Gideon Samid, Computer-Organized Cost Engineering, 2020
The role of an estimator is to find a good way to quantify the five need categories in order to express need dynamics in a scientific way. And in particular to study historical correlations between these need categories and actual cost. It is easy to count dwelling units, or follow on labor statistics. But how do you measure transportation? There are countless detailed models of course, but often their complexity becomes inhibitory and despite their validity and accuracy they turn useless. A cost engineer, unlike a basic scientist can not be satisfied with an accurate quantification method, he or she must mind convenience of use. Level of transportation services in a given area can be roughly but effectively estimated through selecting randomly (or semi-randomly) two points on the area map, and then establishing (1) the direct distance between the two points, and (2) the distance along the existing roads. The ratio between the two reflects accessibility. The random selection can be repeated several times, and the average ratio figure will then express a T-level for the area. One can repeat the process with old maps and thereby quantify a trend. Usually the more elaborate, more accurate-in-principle models will not have enough historic data to execute a similar trend analysis. In the distant past, official clerks did not know that a future scientist will design a fancy model that requires all sorts of detailed data. Yet road maps are available throughout the county history. I dwell a bit on this example because of its wide implications: accuracy and usability may go hand in hand, but not necessarily.
Analysis of spatio-temporal variation of hydroclimatic variables of the Krishna river basin under future scenarios
Published in International Journal of River Basin Management, 2022
Chanapathi Tirupathi, Thatikonda. Shashidhar
The trend is a pattern of gradual change in the data points of a series in a certain direction. It may be upward or downward or no change over time. The trend analysis is used to quantify or to find the patterns in a data series over time. In this study, a non-parametric statistical test, known as Mann-Kendall (MK) test is used to assess, if there are any monotonic (upward or downward) trend in data series (Bisht et al., 2018; Kendall, 1975; Kundu et al., 2015; Mann, 1945; Praveenkumar and Jothiprakash, 2020; Zhao et al., 2015). The monotonic trend means a gradual change in a series either upward or downward direction over time. The main advantage of this test is that it is distribution-free and is applicable to all the distributions. The Null hypothesis (H0) considered in this study is there is no trend in the series. The initial assumption is that null hypothesis (H0) is true. The MK test has its own parameters to reject or accept the null hypothesis. Further details of the Mann Kendall (MK) test and Sen’s slope were provided in the supplementary materials (Appendix A. 3).