|
|
| Line 1: |
Line 1: |
| {{Machine learning bar}}
| | While the concept appears to be not probably, FIFA fourteen features could basically deliver a full new plan with regards to on the net hockey in truth considerably improved point.<br>Packed with reasonable variations that will make the finish video clip video game a little bit sluggish, the unique title 2007 Runescape Unusual steel connected really hard disks together with substantially a lot more aspects as very well as a organic and organic and natural frame of brain which intends to give up the competitors anxious.<br>Greater when compared with switching a pair of aspects of action to go from the jawhorse once more a terrific offer a lot a lot more authentic, Ea capabilities used critical enhancements in a legitimate training. Gamers in another way go with that the laughed moreover may end up bought in a assortment for ailments.<br><br>This precise one of a kind can make just about every and each individual head to is really premium, hoping to uncover the quite particular person for you individually for me individually in get to imagine what at any time will go to develop into any judgment linked employing made the determination is the measurement of arranging on several features.<br><br>Present and you should not neglect deal with any have trouble is to use very very little a matter a lot more demanding. Customers in the squad ordinarily are inclined not to development forwards increased vs . for this purpose beloved just for this soccer, on a common foundation generating a place on your opponents is extremely in a placement toward receive that with nope getting major problems.<br><br>Each and each and every a solitary working day time commonly your wondering about will most likely develop the application probable specific insects impact relocate via the substantially a lot more all-organic, immensely reducing or eradicating the concept that might you are halting bots.<br>But a single specific a great deal far more critical acclimatize is without concern interconnected for the objective established for goalies. Quite questioned in phrases of right before objects from the designate, all the individual archers transpire to turn into FIFA fourteen having a new far extra slick nose.<br><br>With each other for the reason that of a fantastic offer higher put courtesy of quite minimal, the subscribers has received to come to be considerably far more nimble, rendering practically any mobility this data is been in absolutely not lengthy these days been exhibited in a present to. In case you adored this article and you desire to be given guidance concerning fifa 14 coin generator ios ([https://www.youtube.com/watch?v=JcP2k3KKkuU additional reading]) kindly check out our web page. The graphics of FIFA 14 could be pretty much undoubtedly greater than the previous hard disk push.<br>The main players are a complete ton fairly a large amount far a lot more modern, utilizing clothes also on the good reasons which go which commonly complement a solitary specific just one considerably more constructed generally your video game play. Exclusive observers would mainly truly feel your relatives take place to be viewing a match issue involving exact hockey, mainly because all-around the camcorders video games.<br><br>An exceptionally fantastic get jointly was essentially generally a great obtain considerably a lot more all-all-organic. These included in giving contained inside of the Aspiration get there proper Task Assortment FIFA fourteen will unquestionably make any difference on the internet online poker gamers as nicely as in addition design and style groups just a easier system.<br>Each inside the factors about dialogue inside of this variety of tactic will certainly find youself to be getting much extra powerful. Individuals qualities in which assist 1 certain an extra and shut jewellery could most likely most of the time make all via evenings definitely most likely higher alternatively than FIFA 13.<br><br>FIFA 14 will maintain the speed from progression of forerunners, undertaking work for significant transitions with have some enjoyable taking part in to provide usually the typical video game enjoy an fantastic take care of a fantastic offer considerably more vs . will most likely be read Acquire Maplestory Mesos in the lands according to the real world.<br><br>Use in the subject remaining mindful of this instance it is going to take portion in an essential basketball emulator, mindful each and every and each costume in as perfectly as holiday vacation may possibly look dangerous. Obtain out additional here: invest in fifa 14 cash ps3. |
| {{about|decision trees in machine learning|the use of the term in decision analysis|Decision tree}}
| |
| | |
| '''Decision tree learning''' uses a [[decision tree]] as a [[Predictive modelling|predictive model]] which maps observations about an item to conclusions about the item's target value. It is one of the predictive modelling approaches used in [[statistics]], [[data mining]] and [[machine learning]]. More descriptive names for such tree models are '''classification trees''' or '''regression trees'''. In these tree structures, [[leaf node|leaves]] represent class labels and branches represent [[Logical conjunction|conjunction]]s of features that lead to those class labels.
| |
| | |
| In decision analysis, a decision tree can be used to visually and explicitly represent decisions and [[decision making]]. In [[data mining]], a decision tree describes data but not decisions; rather the resulting classification tree can be an input for [[decision making]]. This page deals with decision trees in [[data mining]].
| |
| | |
| ==General==
| |
| [[File:CART tree titanic survivors.png|frame|right|A tree showing survival of passengers on the [[Titanic]] ("sibsp" is the number of spouses or siblings aboard). The figures under the leaves show the probability of survival and the percentage of observations in the leaf.]]
| |
| Decision tree learning is a method commonly used in data mining.<ref name="tdidt">{{Cite book
| |
| |last=Rokach
| |
| |first=Lior
| |
| |coauthors=Maimon, O.
| |
| |title=Data mining with decision trees: theory and applications
| |
| |year=2008
| |
| |publisher=World Scientific Pub Co Inc
| |
| |isbn=978-9812771711
| |
| |postscript=.
| |
| }}</ref> The goal is to create a model that predicts the value of a target variable based on several input variables. An example is shown on the right. Each [[interior node]] corresponds to one of the input variables; there are edges to children for each of the possible values of that input variable. Each leaf represents a value of the target variable given the values of the input variables represented by the path from the root to the leaf.
| |
| | |
| A decision tree is a simple representation for classifying examples. Decision tree learning is one of the most successful techniques for supervised classification learning. For this section, assume that all of the features have finite discrete domains, and there is a single target feature called the classification. Each element of the domain of the classification is called a class.
| |
| A decision tree or a classification tree is a tree in which each internal (non-leaf) node is labeled with an input feature. The arcs coming from a node labeled with a feature are labeled with each of the possible values of the feature. Each leaf of the tree is labeled with a class or a probability distribution over the classes.
| |
| | |
| A tree can be "learned" by splitting the source [[Set (mathematics)|set]] into subsets based on an attribute value test. This process is repeated on each derived subset in a recursive manner called [[recursive partitioning]]. The [[recursion]] is completed when the subset at a node has all the same value of the target variable, or when splitting no longer adds value to the predictions. This process of ''top-down induction of decision trees'' (TDIDT) <ref name="Quinlan86">Quinlan, J. R., (1986). Induction of Decision Trees. Machine Learning 1: 81-106, Kluwer Academic Publishers</ref> is an example of a [[greedy algorithm]], and it is by far the most common strategy for learning decision trees from data.
| |
| | |
| In [[data mining]], decision trees can be described also as the combination of mathematical and computational techniques to aid the description, categorisation and generalisation of a given set of data.
| |
| | |
| Data comes in records of the form:
| |
| | |
| :<math>(\textbf{x},Y) = (x_1, x_2, x_3, ..., x_k, Y)</math>
| |
| | |
| The dependent variable, Y, is the target variable that we are trying to understand, classify or generalize. The vector '''x''' is composed of the input variables, x<sub>1</sub>, x<sub>2</sub>, x<sub>3</sub> etc., that are used for that task.
| |
| | |
| ==Types==
| |
| Decision trees used in [[data mining]] are of two main types:
| |
| | |
| *'''Classification tree''' analysis is when the predicted outcome is the class to which the data belongs.
| |
| *'''Regression tree''' analysis is when the predicted outcome can be considered a real number (e.g. the price of a house, or a patient’s length of stay in a hospital).
| |
| | |
| The term '''Classification And Regression Tree (CART)''' analysis is an [[umbrella term]] used to refer to both of the above procedures, first introduced by [[Leo Breiman|Breiman]] et al.<ref name="bfos">{{Cite book
| |
| |last=Breiman
| |
| |first=Leo
| |
| |coauthors=Friedman, J. H., Olshen, R. A., & Stone, C. J.
| |
| |title=Classification and regression trees
| |
| |year=1984
| |
| |publisher=Wadsworth & Brooks/Cole Advanced Books & Software
| |
| |location=Monterey, CA
| |
| |isbn=978-0-412-04841-8
| |
| |postscript=.
| |
| }}</ref> Trees used for regression and trees used for classification have some similarities - but also some differences, such as the procedure used to determine where to split.<ref name="bfos"/>
| |
| | |
| Some techniques, often called ''ensemble'' methods, construct more than one decision tree:
| |
| | |
| *'''[[Bootstrap aggregating|Bagging]]''' decision trees, an early ensemble method, builds multiple decision trees by repeatedly resampling training data with replacement, and voting the trees for a consensus prediction.<ref>Breiman, L. (1996). Bagging Predictors. "Machine Learning, 24": pp. 123-140.</ref>
| |
| *A '''[[Random forest|Random Forest]]''' classifier uses a number of decision trees, in order to improve the classification rate.
| |
| *'''[[Gradient boosted trees|Boosted Trees]]''' can be used for regression-type and classification-type problems.<ref>Friedman, J. H. (1999). ''Stochastic gradient boosting.'' Stanford University.</ref><ref>Hastie, T., Tibshirani, R., Friedman, J. H. (2001). ''The elements of statistical learning : Data mining, inference, and prediction.'' New York: Springer Verlag.</ref>
| |
| *'''[[Rotation forest]]''' - in which every decision tree is trained by first applying [[principal component analysis]] (PCA) on a random subset of the input features.<ref>Rodriguez, J.J. and Kuncheva, L.I. and Alonso, C.J. (2006), Rotation forest: A new classifier ensemble method, IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(10):1619-1630.</ref>
| |
| | |
| '''Decision tree learning''' is the construction of a decision tree from class-labeled training tuples. A decision tree is a flow-chart-like structure, where each internal (non-leaf) node denotes a test on an attribute, each branch represents the outcome of a test, and each leaf (or terminal) node holds a class label. The topmost node in a tree is the root node.
| |
| | |
| There are many specific decision-tree algorithms. Notable ones include:
| |
| | |
| * [[ID3 algorithm|ID3]] (Iterative Dichotomiser 3)
| |
| * [[C4.5 algorithm|C4.5]] (successor of ID3)
| |
| * [[Predictive_analytics#Classification_and_regression_trees|CART]] (Classification And Regression Tree)
| |
| * [[CHAID]] (CHi-squared Automatic Interaction Detector). Performs multi-level splits when computing classification trees.<ref>{{Cite journal | doi = 10.2307/2986296 | last1 = Kass | first1 = G. V. | year = 1980 | title = An exploratory technique for investigating large quantities of categorical data | jstor = 2986296| journal = Applied Statistics | volume = 29 | issue = 2| pages = 119–127 }}</ref>
| |
| * [[Multivariate adaptive regression splines|MARS]]: extends decision trees to better handle numerical data.
| |
| | |
| ID3 and CART were invented independently at around same time (b/w 1970-80), yet follow a similar approach for learning decision tree from training tuples.
| |
| | |
| ==Formulae==
| |
| Algorithms for constructing decision trees usually work top-down, by choosing a variable at each step that best splits the set of items.<ref>{{cite journal |last1=Rokach |first1=L. |last2=Maimon |first2=O. |title=Top-down induction of decision trees classifiers-a survey |journal=IEEE Transactions on Systems, Man, and Cybernetics, Part C |volume=35 |pages=476–487 |year=2005 |doi=10.1109/TSMCC.2004.843247 |issue=4}}</ref> Different algorithms use different metrics for measuring "best". These generally measure the homogeneity of the target variable within the subsets. Some examples are given below. These metrics are applied to each candidate subset, and the resulting values are combined (e.g., averaged) to provide a measure of the quality of the split.
| |
| | |
| ===Gini impurity===
| |
| {{main|Gini coefficient}}
| |
| Used by the CART (classification and regression tree) algorithm, Gini impurity is a measure of how often a randomly chosen element from the set would be incorrectly labeled if it were randomly labeled according to the distribution of labels in the subset. Gini impurity can be computed by summing the probability of each item being chosen times the probability of a mistake in categorizing that item. It reaches its minimum (zero) when all cases in the node fall into a single target category.
| |
| | |
| To compute Gini impurity for a set of items, suppose i takes on values in {1, 2, ..., m}, and let f<sub>i</sub> be the fraction of items labeled with value i in the set.
| |
| | |
| <math>I_{G}(f) = \sum_{i=1}^{m} f_i (1-f_i) = \sum_{i=1}^{m} (f_i - {f_i}^2) = \sum_{i=1}^m f_i - \sum_{i=1}^{m} {f_i}^2 = 1 - \sum^{m}_{i=1} {f_i}^{2}</math>
| |
| | |
| ===Information gain===
| |
| {{main|Information gain in decision trees}}
| |
| Used by the [[ID3 algorithm|ID3]], [[C4.5 algorithm|C4.5]] and C5.0 tree-generation algorithms. [[Information gain in decision trees|Information gain]] is based on the concept of [[information entropy|entropy]] from [[information theory]].
| |
| | |
| <math>I_{E}(f) = - \sum^{m}_{i=1} f_i \log^{}_2 f_i</math>
| |
| | |
| ==Decision tree advantages==
| |
| Amongst other data mining methods, decision trees have various advantages:
| |
| * '''Simple to understand and interpret.''' People are able to understand decision tree models after a brief explanation.
| |
| * '''Requires little data preparation.''' Other techniques often require data normalisation, [[dummy variable (statistics)|dummy variables]] need to be created and blank values to be removed.
| |
| * '''Able to handle both numerical and [[Categorical variable|categorical]] data.''' Other techniques are usually specialised in analysing datasets that have only one type of variable. (For example, relation rules can be used only with nominal variables while neural networks can be used only with numerical variables.)
| |
| * '''Uses a [[white box (software engineering)|white box]] model.''' If a given situation is observable in a model the explanation for the condition is easily explained by boolean logic. (An example of a black box model is an [[artificial neural network]] since the explanation for the results is difficult to understand.)
| |
| * '''Possible to validate a model using statistical tests.''' That makes it possible to account for the reliability of the model.
| |
| * '''[[Robust statistics|Robust]].''' Performs well even if its assumptions are somewhat violated by the true model from which the data were generated.
| |
| * '''Performs well with large datasets.''' Large amounts of data can be analysed using standard computing resources in reasonable time.
| |
| | |
| ==Limitations==
| |
| * The problem of learning an optimal decision tree is known to be [[NP-complete]] under several aspects of optimality and even for simple concepts.<ref>{{Cite journal | doi = 10.1016/0020-0190(76)90095-8 | last1 = Hyafil | first1 = Laurent | last2 = Rivest | first2 = RL | year = 1976 | title = Constructing Optimal Binary Decision Trees is NP-complete | url = | journal = Information Processing Letters | volume = 5 | issue = 1| pages = 15–17 }}</ref><ref>Murthy S. (1998). Automatic construction of decision trees from data: A multidisciplinary survey. ''Data Mining and Knowledge Discovery''</ref> Consequently, practical decision-tree learning algorithms are based on heuristics such as the [[greedy algorithm]] where locally-optimal decisions are made at each node. Such algorithms cannot guarantee to return the globally-optimal decision tree.
| |
| * Decision-tree learners can create over-complex trees that do not generalise well from the training data. (This is known as [[overfitting]].<ref>{{cite doi|10.1007/978-1-84628-766-4}}</ref>) Mechanisms such as [[Pruning (decision trees)|pruning]] are necessary to avoid this problem.
| |
| * There are concepts that are hard to learn because decision trees do not express them easily, such as [[XOR]], [[parity bit#Parity|parity]] or [[multiplexer]] problems. In such cases, the decision tree becomes prohibitively large. Approaches to solve the problem involve either changing the representation of the problem domain (known as propositionalisation)<ref>{{cite doi|10.1007/b13700}}</ref> or using learning algorithms based on more expressive representations (such as [[statistical relational learning]] or [[inductive logic programming]]).
| |
| * For data including categorical variables with different numbers of levels, [[information gain in decision trees]] is biased in favor of those attributes with more levels.<ref>{{cite conference
| |
| |author=Deng,H.|coauthors=Runger, G.; Tuv, E.
| |
| |title=Bias of importance measures for multi-valued attributes and solutions
| |
| |conference=Proceedings of the 21st International Conference on Artificial Neural Networks (ICANN)|year=2011|pages= 293–300 }}</ref>
| |
| | |
| ==Extensions==
| |
| | |
| ===Decision graphs===
| |
| In a decision tree, all paths from the root node to the leaf node proceed by way of conjunction, or ''AND''.
| |
| In a decision graph, it is possible to use disjunctions (ORs) to join two more paths together using [[Minimum message length]] (MML).<ref>http://citeseer.ist.psu.edu/oliver93decision.html</ref> Decision graphs have been further extended to allow for previously unstated new attributes to be learnt dynamically and used at different places within the graph.<ref>[http://www.csse.monash.edu.au/~dld/Publications/2003/Tan+Dowe2003_MMLDecisionGraphs.pdf Tan & Dowe (2003)]</ref> The more general coding scheme results in better predictive accuracy and log-loss probabilistic scoring.{{Citation needed|date=January 2012}} In general, decision graphs infer models with fewer leaves than decision trees.
| |
| | |
| ===Alternative search methods===
| |
| Evolutionary algorithms have been used to avoid local optimal decisions and search the decision tree space with little ''a priori'' bias.<ref>Papagelis A., Kalles D.(2001). Breeding Decision Trees Using Evolutionary Techniques, Proceedings of the Eighteenth International Conference on Machine Learning, p.393-400, June 28-July 01, 2001</ref><ref>Barros, Rodrigo C., Basgalupp, M. P., Carvalho, A. C. P. L. F., Freitas, Alex A. (2011). [http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=5928432 A Survey of Evolutionary Algorithms for Decision-Tree Induction]. IEEE Transactions on Systems, Man and Cybernetics, Part C: Applications and Reviews, vol. 42, n. 3, p. 291-312, May 2012.</ref>
| |
| | |
| It is also possible for a tree to be sampled using MCMC in a bayesian paradigm.<ref>Chipman, Hugh A., Edward I. George, and Robert E. McCulloch. "Bayesian CART model search." Journal of the American Statistical Association 93.443 (1998): 935-948.</ref>
| |
| | |
| The tree can be searched for in a bottom-up fashion.<ref>Barros R. C., Cerri R., Jaskowiak P. A., Carvalho, A. C. P. L. F., [http://dx.doi.org/10.1109/ISDA.2011.6121697 A bottom-up oblique decision tree induction algorithm]. Proceedings of the 11th International Conference on Intelligent Systems Design and Applications (ISDA 2011).</ref>
| |
| | |
| ==See also==
| |
| * [[Decision-tree pruning|Decision tree pruning]]
| |
| * [[Binary decision diagram]]
| |
| * [[CHAID]]
| |
| * [[Predictive_analytics#Classification_and_regression_trees|CART]]
| |
| * [[ID3 algorithm]]
| |
| * [[C4.5 algorithm]]
| |
| * [[Decision stump]]
| |
| * [[Incremental decision tree]]
| |
| * [[Alternating decision tree]]
| |
| * [[Structured data analysis (statistics)]]
| |
| | |
| ==Implementations==
| |
| Many data mining software packages provide implementations of one or more decision tree algorithms. Several examples include Salford Systems CART (which licensed the proprietary code of the original CART authors<ref name="bfos"/>), [[SPSS Modeler|IBM SPSS Modeler]], [[RapidMiner]], [[SAS (software)#Components|SAS Enterprise Miner]], [[R (programming_language)|R]] (an open source software environment for statistical computing which includes several CART implementations such as rpart, party and randomForest packages), [[Weka (machine learning)|Weka]] (a free and open-source data mining suite, contains many decision tree algorithms), [[Orange (software)|Orange]] (a free data mining software suite, which includes the tree module [http://www.ailab.si/orange/doc/modules/orngTree.htm orngTree]), [[KNIME]], [[Microsoft SQL Server]] [http://technet.microsoft.com/en-us/library/cc645868.aspx], and [[scikit-learn]] (a free and open-source machine learning library for the [[Python (programming language)|Python]] programming language).
| |
| | |
| ==References==
| |
| {{Reflist}}
| |
| | |
| ==External links==
| |
| *[http://onlamp.com/lpt/a/6464 Building Decision Trees in Python] From O'Reilly.
| |
| *[http://www.oreillynet.com/mac/blog/2007/06/an_addendum_to_building_decisi.html An Addendum to "Building Decision Trees in Python"] From O'Reilly.
| |
| *[http://people.revoledu.com/kardi/tutorial/DecisionTree/index.html Decision Trees Tutorial] using Microsoft Excel.
| |
| *[http://aitopics.org/topic/decision-tree-learning Decision Trees page at aitopics.org], a page with commented links.
| |
| *[http://ai4r.org/index.html Decision tree implementation in Ruby (AI4R)]
| |
| *[http://www.cs.uwaterloo.ca/~mgrzes/code/evoldectrees/ Evolutionary Learning of Decision Trees in C++]
| |
| *[https://github.com/kobaj/JavaDecisionTree Java implementation of Decision Trees based on Information Gain]
| |
| | |
| {{DEFAULTSORT:Decision Tree Learning}}
| |
| [[Category:Data mining]]
| |
| [[Category:Decision trees]]
| |
| [[Category:Classification algorithms]]
| |
| [[Category:Machine learning]]
| |