Violation of independence assumption in ICA and its consequences

2021 ◽  
pp. 187-194
Author(s):  
Bhaveshkumar C. Dharmani
Keyword(s):  
Independence Assumption

Following Leaders

2018 ◽  
Author(s):  
Robert E. Goodin ◽  
Kai Spiekermann
Keyword(s):  
Democratic Theory ◽  
Opinion Leaders ◽  
Independence Assumption ◽  
Positive Correlation ◽  
The Third ◽  
Single Opinion

The question of leadership is connected to many central debates in democratic theory. In this chapter, the focus is on leadership in terms of beliefs, not desires. Opinion leaders’ influence undermines the Independence Assumption. The first section looks at single opinion leaders, who, if their influence is strong and their competence limited, reduce group competence, often severely. The second section considers multiple correlated opinion leaders. The effects depend on the negative or positive correlation between the opinion leaders, the number of voters following each, and the competence of leaders. Multiple uncorrelated opinion leaders are the topic of the third section. Their influence can be relatively benign if they are many and if they are reasonably competent. Finally, a great many ‘local’ opinion leaders, as envisaged by Lazarsfeld, Berelson, and Gaudet, can offset the negative epistemic impact of a few ‘big’ opinion leaders.

scholarly journals Structure Extension of Tree-Augmented Naive Bayes

Entropy ◽  
2019 ◽  
Vol 21 (8) ◽  
pp. 721 ◽  
Author(s):  
YuGuang Long ◽  
LiMin Wang ◽  
MingHui Sun
Keyword(s):  
Naive Bayes ◽  
Classification Performance ◽  
Naïve Bayes ◽  
Training Data ◽  
Conditional Probability Distribution ◽  
Independence Assumption ◽  
Bayesian Network Classifiers ◽  
Leibler Divergence ◽  
The Difference ◽  
Structure Extension

Due to the simplicity and competitive classification performance of the naive Bayes (NB), researchers have proposed many approaches to improve NB by weakening its attribute independence assumption. Through the theoretical analysis of Kullback–Leibler divergence, the difference between NB and its variations lies in different orders of conditional mutual information represented by these augmenting edges in the tree-shaped network structure. In this paper, we propose to relax the independence assumption by further generalizing tree-augmented naive Bayes (TAN) from 1-dependence Bayesian network classifiers (BNC) to arbitrary k-dependence. Sub-models of TAN that are built to respectively represent specific conditional dependence relationships may “best match” the conditional probability distribution over the training data. Extensive experimental results reveal that the proposed algorithm achieves bias-variance trade-off and substantially better generalization performance than state-of-the-art classifiers such as logistic regression.

scholarly journals Severe violations of independence in response inhibition tasks

2021 ◽  
Vol 7 (12) ◽  
pp. eabf4355
Author(s):  
Patrick G. Bissett ◽  
Henry M. Jones ◽  
Russell A. Poldrack ◽  
Gordon D. Logan
Keyword(s):  
Response Inhibition ◽  
Theoretical Foundation ◽  
Reaction Times ◽  
Stop Signal ◽  
Experimental Paradigm ◽  
Signal Delay ◽  
Independence Assumption ◽  
Analysis Techniques ◽  
Wide Range ◽  
Saccadic Responses

The stop-signal paradigm, a primary experimental paradigm for understanding cognitive control and response inhibition, rests upon the theoretical foundation of race models, which assume that a go process races independently against a stop process that occurs after a stop-signal delay (SSD). We show that severe violations of this independence assumption at short SSDs occur systematically across a wide range of conditions, including fast and slow reaction times, auditory and visual stop signals, manual and saccadic responses, and especially in selective stopping. We also reanalyze existing data and show that conclusions can change when short SSDs are excluded. Last, we suggest experimental and analysis techniques to address this violation, and propose adjustments to extant models to accommodate this finding.

scholarly journals Impacts of ditch cleaning on hydrological processes in a drained peatland forest

2008 ◽  
Vol 12 (5) ◽  
pp. 1211-1227 ◽  
Author(s):  
H. Koivusalo ◽  
E. Ahti ◽  
A. Laurén ◽  
T. Kokkonen ◽  
T. Karvonen ◽  
...  
Keyword(s):  
Numerical Experiment ◽  
Water Table ◽  
Mineral Soil ◽  
Annual Runoff ◽  
Independence Assumption ◽  
Peatland Hydrology ◽  
Network Maintenance ◽  
And Control ◽  
Peat Formation ◽  
Ditch Cleaning

Abstract. One fourth of the forests in Finland are growing on drained peatlands. Forestry operations such as ditch network maintenance increase the export of suspended solids and nutrients, and deteriorate water quality in lakes and rivers. Water protection presupposes an understanding of how forestry operations affect peatland hydrology. The objective was to study the hydrological impacts of ditch cleaning on the basis of water table level and runoff measurements from two pairs of artificially delineated catchments in drained peatland forests in Finland. Data from treated and control catchments indicated that ditch cleaning lowered the level of the water table in sites where a shallow peat layer was underlain by mineral soil. In sites with deep peat formation, the water table showed no detectable response to ditch cleaning. Runoff data suggested that annual runoff clearly increased after ditch cleaning, which was in conflict with the previously reported small impacts of ditch network maintenance. The hydrological model FEMMA was calibrated and applied to assess the conformity of the data and the experimental setup. In the model application, the catchments were assumed to behave as independent hydrological units. However, assessment of the model results and the measurements suggested that ditch cleaning had an impact on hydrological measurements in both treated and control catchments. It appeared that the independence assumption was violated and there was a hydrological connection between the artificial catchments and, therefore, the results of the data analysis were considered misleading. Finally, a numerical experiment based on the model simulations was conducted to explain how the assumed relationship between soil moisture and transpiration is reflected in the modelled runoff. Modelled runoff decreases and evaporation increases when ditches are cleaned in poorly drained sites, where the initial ditch depth is small and the depth of a highly conductive topsoil layer is low. The numerical experiment can be applied to assess when ditch cleaning does not improve evapotranspiration and is unnecessary.

Improving the Naïve Bayes Classifier

2011 ◽  
pp. 879-883 ◽  
Author(s):  
Liwei Fan ◽  
Kim Leng Poh
Keyword(s):  
Naive Bayes ◽  
Probability Distributions ◽  
Naïve Bayes ◽  
Knowledge Representation And Reasoning ◽  
Naive Bayes Classifier ◽  
Bayes Classifier ◽  
Naïve Bayes Classifier ◽  
Real World Data ◽  
Independence Assumption ◽  
Strong Assumption

A Bayesian Network (BN) takes a relationship between graphs and probability distributions. In the past, BN was mainly used for knowledge representation and reasoning. Recent years have seen numerous successful applications of BN in classification, among which the Naïve Bayes classifier was found to be surprisingly effective in spite of its simple mechanism (Langley, Iba & Thompson, 1992). It is built upon the strong assumption that different attributes are independent with each other. Despite of its many advantages, a major limitation of using the Naïve Bayes classifier is that the real-world data may not always satisfy the independence assumption among attributes. This strong assumption could make the prediction accuracy of the Naïve Bayes classifier highly sensitive to the correlated attributes. To overcome the limitation, many approaches have been developed to improve the performance of the Naïve Bayes classifier. This article gives a brief introduction to the approaches which attempt to relax the independence assumption among attributes or use certain pre-processing procedures to make the attributes as independent with each other as possible. Previous theoretical and empirical results have shown that the performance of the Naïve Bayes classifier can be improved significantly by using these approaches, while the computational complexity will also increase to a certain extent.

Sign in / Sign up

Export Citation Format

Share Document