An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression

Altman, Naomi

doi:10.1080/00031305.1992.10475879

Cited by 3,498 publications

(815 citation statements)

References 29 publications

Supporting

Mentioning

803

Contrasting

Unclassified

Order By: Relevance

“…These algorithms were chosen based on their potential ability to determine the relationship between inputs (V WE , V AE , T ) and outputs ([SO 2 ]) without needing to know the functional form of the relationship itself. The methods examined were: ridge regression (RR), which attempts to reduce standard error by introducing bias to reduce multicollinearity among independent variables (Rifkin, 2007); least absolute shrinkage and selection operator (LASSO) regression, which similarly reduces covariance and overfitting by eliminating similar features and imposing an absolute limit on the sum of the coefficients (Tibshirani, 1996); classification and decision trees (CART), which forms a collection of rules based in a recursive fashion by selecting data that differentiate observations based on the dependent variable (Breiman et al, 1984); and k nearest neighbors regression (kNN), which estimates the regression curve without making assumptions about the structure of the model (Altman, 1992). The kNN approach, which was found to have the best performance (see Results, below), involves mapping input variables from the training data (V WE , V AE , T ) to the output variable (SO 2 mixing ratio) in an ndimensional vector space.…”

Section: Nonparametric Calibration Approachesmentioning

confidence: 99%

Calibration and assessment of electrochemical air quality sensors by co-location with regulatory-grade instruments

et al. 2018

View full text Add to dashboard Cite

Abstract. The use of low-cost air quality sensors for air pollution research has outpaced our understanding of their capabilities and limitations under real-world conditions, and there is thus a critical need for understanding and optimizing the performance of such sensors in the field. Here we describe the deployment, calibration, and evaluation of electrochemical sensors on the island of Hawai'i, which is an ideal test bed for characterizing such sensors due to its large and variable sulfur dioxide (SO 2 ) levels and lack of other co-pollutants. Nine custom-built SO 2 sensors were colocated with two Hawaii Department of Health Air Quality stations over the course of 5 months, enabling comparison of sensor output with regulatory-grade instruments under a range of realistic environmental conditions. Calibration using a nonparametric algorithm (k nearest neighbors) was found to have excellent performance (RMSE < 7 ppb, MAE < 4 ppb, r 2 > 0.997) across a wide dynamic range in SO 2 (< 1 ppb, > 2 ppm). However, since nonparametric algorithms generally cannot extrapolate to conditions beyond those outside the training set, we introduce a new hybrid linear-nonparametric algorithm, enabling accurate measurements even when pollutant levels are higher than encountered during calibration. We find no significant change in instrument sensitivity toward SO 2 after 18 weeks and demonstrate that calibration accuracy remains high when a sensor is calibrated at one location and then moved to another. The performance of electrochemical SO 2 sensors is also strong at lower SO 2 mixing ratios (< 25 ppb), for which they exhibit an error of less than 2.5 ppb. While some specific results of this study (calibration accuracy, performance of the various algorithms, etc.) may differ for measurements of other pollutant species in other areas (e.g., polluted urban regions), the calibration and validation approaches described here should be widely applicable to a range of pollutants, sensors, and environments.

show abstract

Section: Nonparametric Calibration Approachesmentioning

confidence: 99%

Calibration and assessment of electrochemical air quality sensors by co-location with regulatory-grade instruments

et al. 2018

View full text Add to dashboard Cite

show abstract

“…In example acknowledgment, the k-closest neighbors calculation (k-NN) is a non-parametric technique utilized for order and relapse. In both cases, the info comprises of the k nearest preparing cases in the component space [3]. The yield relies on upon whether k-NN is utilized for arrangement or relapse:…”

Section: E Genetic Algorithmsmentioning

confidence: 99%

“…The tenets are checked, and the ones that fit the information best are kept, the guidelines that don't fit the information are discarded. [3] The principles that were kept are then changed, and duplicated to make new rules. This procedure repeats as essential with a specific end goal to deliver decide that matches the dataset as nearly as possible.…”

Section: E Genetic Algorithmsmentioning

confidence: 99%

A Survey on Data Mining

Kumar¹

2017

IJRASET

View full text Add to dashboard Cite

Information mining is the figuring procedure of finding examples in expansive informational collections including techniques at the convergence of counterfeit consciousness, machine learning, measurements, and database frameworks. It is an interdisciplinary subfield of software engineering. The general objective of the information mining procedure is to concentrate data from an informational index and change it into a justifiable structure for further utilize. In this paper, the basic concept of data mining is being discussed along with the various types of techniques exist. The data mining tools are also described in this study to help researchers in using them foe enhancing and collecting various patterns and relationships from huge data. Keywords: Mining; genetic; weka; neural; tools; I.INTRODUCTION The measure of information being created and put away in expanding by a quick speed because of progression of new innovations .With heaps of information comes awesome obligations of taking care of it so it can be utilized when required .In our exploration paper we will talk about on how information mining helps us in dealing with this information and removing data when required. Information mining is the way toward separating data from an informational index and changes it according to our requirements. Around 1960's another procedure was developed named as information angling or information digging. Notwithstanding, it was considered as an awful practice since it investigated information with no earlier theory. In late 1980's, this term database mining was advanced from information angling by HNC, a san diego based organization. Information mining term appeared after the global meeting on information mining and learning revelation (KDD-95) held in Montreal under AAAI sponsorship. At long last in 1996, Usama Fayyad propelled information mining and learning revelation. There are different strides for information mining. The initial three stages include organizing or sorting the data and taking out the one we require from the entire part of information. At that point, the information must be changed into an appropriate way to such an extent that all the essential calculations can be connected to this. For instance information change is given by Cortes and Pregibon (1998). In the event that every information record depicts one telephone call however the objective is to foresee whether a telephone number has a place with a business or private client in light of its calling designs, then all records related with each telephone number must be accumulated, which will involve making ascribes comparing to the normal number of calls every day, normal call span, and so on. Each one of those inspired by this point ought to peruse the book Pyle 1999. The fourth step principally concentrates on the change of this information into information into trees by means of calculations to discover the examples. The term Knowledge Discovery in Databases, or KDD for short, alludes to the expansive procedure of discovering infor...

show abstract

“…ML does not make any assumptions about the right structure of the data model, allowing the construction of complex non-linear models. There are many different paradigms in ML: lazy methods such as K-Nearest Neighbours (KNN) (Altman, 1992) methods, based on tree construction, as, for instance, C4.5 (Quinlan, 1993) or Neural or Bayesian networks (Mitchell, 1997). All of them have been successfully used in many different domains.…”

Section: Introductionmentioning

confidence: 99%

“…When the variable to predict is continuous, ML methods more commonly used are CART (Breiman, 2001), M5 (Quinlan, 1992), M5-Prime (Wang & Witten, 1997), KNN (Altman, 1992), or support vector regression (SVR, see Basak et al, 2007).…”

Section: Introductionmentioning

confidence: 99%

Machine learning applied to the prediction of citrus production

Dı́az

Mazza

Combarro

et al. 2017

Span. j. agric. res.

View full text Add to dashboard Cite

An in-depth knowledge about variables affecting production is required in order to predict global production and take decisions in agriculture. Machine learning is a technique used in agricultural planning and precision agriculture. This work (i) studies the effectiveness of machine learning techniques for predicting orchards production; and (ii) variables affecting this production were also identified. Data from 964 orchards of lemon, mandarin, and orange in Corrientes, Argentina are analysed. Graphic and analytical descriptive statistics, correlation coefficients, principal component analysis and Biplot were performed. Production was predicted via M5-Prime, a model regression tree constructor which produces a classification based on piecewise linear functions. For all the species studied, the most informative variable was the trees' age; in mandarin and orange orchards, age was followed by between and within row distances; irrigation also affected mandarin production. Also, the performance of M5-Prime in the prediction of production is adequate, as shown when measured with correlation coefficients (~0.8) and relative mean absolute error (~0.1). These results show that M5-Prime is an appropriate method to classify citrus orchards according to production and, in addition, it allows for identifying the most informative variables affecting production by tree.

show abstract

An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression

Cited by 3,498 publications

References 29 publications

Calibration and assessment of electrochemical air quality sensors by co-location with regulatory-grade instruments

Calibration and assessment of electrochemical air quality sensors by co-location with regulatory-grade instruments

A Survey on Data Mining

Machine learning applied to the prediction of citrus production

Contact Info

Product

Resources

About