Data mining with agricultural soil databases is a relatively young research area. In agricultural field, the determination of soil category mainly depends on the atmospheric conditions and different soil characteristics. Classification as an essential data mining technique used to develop models describing different soil classes. Such analysis can present us with a complete understanding of various soil databases at large. In our study, we proposed a novel Neuro-fuzzy classification based technique and applied it to large soil databases to find out significant relationships. We used our technique to three benchmark data sets from the UCI machine learning repository for soil categorization and they were namely Statlog (Landsat Satellite), Covertype, and 3 data sets. Our objective was to develop an efficient classification model with the proposed method and, therefore compare its performance with two well-known supervised classification algorithms Multilayer Perceptron and Support Vector Machine. We estimated the performance of these classification techniques in terms of different evaluation measures like Accuracy, Kappa statistic, True-Positive Rate, False-Positive Rate, Precision, Recall, and F-Measure. The proposed technique had an accuracy of 99.4 % with the Statlog data set, 97.7 % with the Covertype data set and 90 % with the 3 data set; and in every aspect, it performed better than Multilayer Perceptron and Support Vector Machine algorithms.
Data mining consists of extracting interesting patterns representing knowledge from real-world databases. The software applications related with data mining includes various methodologies developed by both commercial and research organizations. Different data mining techniques used to...
... middle of paper ...
...It combines the human alike logical reasoning of fuzzy based systems with the learning and connectedness structure of ANNs by means of the fuzzy sets and linguistic model based approaches. In our work, we proposed a novel Neuro-fuzzy based classification method for soil data mining. We applied our method to three benchmark data sets from the UCI machine learning repository for soil classification and, therefore compare its performance with MLP and SVM based classification models.
This research study is arranged as follows: Section 2 includes the related works done in this field; Section 3 describes our proposed Neuro-fuzzy classification based method. Section 4 explains the methodology in terms of our proposed neuro-fuzzy method, MLP, and SVM. Section 5 discusses the classification performance analysis and results; and the Section 6 is reserved for the conclusion.
Stephen V. Stehman, “Selecting and interpreting measures of thematic classification accuracy”. Remote Sensing of Environment, Vol. 62, No.1, pp.77–89, 1997.
In particular, I have special interest in focusing in Agriculture, due to my home region mainly has an agricultural profile. During my training I realized the importance of reliable and quality information sources. Similarly, I recognize in satellite and aerial imagery a rich source of information. Specifically, in the future I would like to exploit this type of data for the study of soil quality and crop performance in order to unveil patterns that allow us to better understand their features and shortcomings.
Traditional business intelligence tools are being replaced by data discovery software. The data discovery software has numerous capabilities that are dominating purchase requirements for larger distribution. A challenge remaining is the ability to meet the dual demands of enterprise IT and business users.
Classification Text documents are arranged into groups of pre-labeled class. Learning schemes learn through training text documents and efficiency of these system is tested by using test text documents. Common algorithms include decision tree learning, naive Bayesian classification, nearest neighbor and neural network. This is called supervised learning.
Stergiou, C., & Siganos, D. (2011, August 6). Neural Networks. Retrieved August 6, 2011, from
Key Words; Artificial Intelligence, Multiple Intelligence, Fuzzy Logic, Fuzzy Logic Toolbox, Vocational Guidance, Decision Making
Data mining has emerged as an important method to discover useful information, hidden patterns or rules from different types of datasets. Association rule mining is one of the dominating data mining technologies. Association rule mining is a process for finding associations or relations between data items or attributes in large datasets. Association rule is one of the most popular techniques and an important research issue in the area of data mining and knowledge discovery for many different purposes such as data analysis, decision support, patterns or correlations discovery on different types of datasets. Association rule mining has been proven to be a successful technique for extracting useful information from large datasets. Various algorithms or models were developed many of which have been applied in various application domains that include telecommunication networks, market analysis, risk management, inventory control and many others
A data stream is a real time, continuous, structured sequence of data items. Mining data stream is the process of extracting knowledge from continuous, rapid data records. Data arrives faster, so it is a very difficult task to mine that data. Stream mining algorithms typically need to be designed so that the algorithm works with one pass of the data. Data streams are a computational challenge to data mining problems because of the additional algorithmic constraints created by the large volume of data. In addition, the problem of temporal locality leads to a number of unique mining challenges in the data stream case. The data mining techniques namely clustering, classification and frequent pattern mining are applied to extract the knowledge from the data streams. This research work mainly concentrates on how to find the valuable items found in a transactional data of a data stream. In the literature, most of the researchers have discussed about how the frequent items are mined from the data streams. This research work helps to find the valuable items in a transactional data. This is a new research idea in the area of data stream frequent pattern mining. Frequent Item mining is defined as finding the items which are occurring frequently and above the given threshold. Valuable item is nothing but finding the costliest item or most valuable items in a data base. Predicting this information helps businesses to know about the sales details about the valuable items which guide to make important decisions, such as catalogue drawing, cross marketing, consumer shopping and performance scrutiny. In this research work, two new algorithms namely VIM (Valuable Item Mining) and TVIM (Tree based Valuable Item Mining) are proposed for finding the...
The fuzzy is basic set of rules which is based on system error and change in error which expert advice into automatic control condition for self adaptive controller. Fuzzy represents a sequence of control mechanism to adjust the effect of certain system stimulations. It reflects the expert conditions in to appropriate control design.
HAND, D. J., MANNILA, H., & SMYTH, P. (2001).Principles of data mining. Cambridge, Mass, MIT Press.
... applied on different Domain data sets and sub level data sets. The data sets are applied on Maximum entropy, Support Vector Machine Method, Multinomial naïve bayes algorithms, I got 60-70% of accuracy. The above is also applied for the Unigrams of Maximum entropy, Support Vector Machine Method, Multinomial naïve bayes algorithms achieved an accuracy of 65-75%. Applied the same data on proposed lexicon Based Semantic Orientation Analysis Algorithm, we received better accuracy of 85%. In subjective Feature Relation Networks Chi-square model using n-grams, POS tagging by applying linguistic rules performed with highest accuracy of 80% to 93% significantly better than traditional naïve bayes with unigram model. The after applying proposed model on different sets the results are validated with test data and proved our methods are more accurate than the other methods.
The dynamics of our society bring many challenges and opportunities to the business world. Within the last decade, hundreds of jobs have emerged particularly in the technology sector to help keep up with the ever-changing world and to compete on a larger and better scale than the competition. Two key job markets and the basis of this research paper are business intelligence or BI and data mining or DM. These two fields play a very important role in small to large companies and are becoming higher desired sectors within the back offices of the workplace. This paper will explore what the meaning of BI and DM really is, how they are used and what we can expect as workers and learners of the technology and business fields for the future.
Machine learning systems can be categorized according to many different criteria. We will discuss three criteria: Classification on the basis of the underlying learning strategies used, Classification on the basis of the representation of knowledge or skill acquired by the learner and Classification in terms of the application domain of the performance system for which knowledge is acquired.
The texture refers to the structure of the soil in relation to small, medium or large particles in a specific soil mass (Ball 2001). Soil texture is classified based on the amount of sand, silt and clay present in a soil sample (Schoonover & Crim 2015). A coarse soil is a sand or loamy soil, a medium soil is a loam, silt loam or silt whereas a fine soil is a sandy clay, silty clay or just clay (Ball 2001). The particles of the clay are very small which means they have a large surface area (What is Soil Texture? 2017). Due to the surface area, the water gets stuck well to the clay and its ability to retain moisture gets high (What is Soil Texture? 2017). If the surface area is high, more area is available for positively charged plant nutrients
Sammut, S. (N/A). A Soil Information System for the Maltese Islands. Available: http://www.wise-rtd.info/sites/default/files/d-2008-05-26-Project_presentation.pdf. Last accessed 29th Dec 2013.