A Bayesian Network Structure Learning Approach to Identify Genes Associated with Stress in Spleens of Chickens

Abstract Differences in the expression patterns of genes have been used to measure the effects of non-stress or stress conditions in poultry species. However, the list of genes identified can be extensive and they might be related to several biological systems. Therefore, the aim of this study was to identify a small set of genes closely associated with stress in a poultry animal model, the chicken (Gallus gallus), by reusing and combining data previously published together with bioinformatic analysis and Bayesian networks in a multi-step approach. Two datasets were collected from publicly available repositories and pre-processed. Bioinformatics analyses were performed to identify genes common to both datasets that showed differential expression patterns between non-stress and stress conditions. Bayesian networks were learnt using a Simulated Annealing algorithm implemented in the software Banjo. The structure of the Bayesian network consisted of 16 out of 19 genes in addition to the stress condition. CARD19 displayed a direct relationship with the stress condition, and three other genes, CYGB, BRAT1, and EPN3 were also relevant for the stress condition. The biological functionality of these genes are related to damage, apoptosis, and oxygen provision, and they could potentially be further explored as biomarkers of stress.

Download Full-text

Bayesian Networks for Image Understanding

Bayesian Network Technologies ◽

10.4018/978-1-59904-141-4.ch007 ◽

2007 ◽

pp. 128-150

Author(s):

Andreas Savaki ◽

Jiebo Luo ◽

Michael Kane

Keyword(s):

Bayesian Networks ◽

Object Detection ◽

Bayesian Network ◽

Structure Learning ◽

Image Understanding ◽

Semantic Features ◽

Scene Classification ◽

Low Level ◽

Object Parts ◽

Scene Content

Image understanding deals with extracting and interpreting scene content for use in various applications. In this chapter, we illustrate that Bayesian networks are particularly well-suited for image understanding problems, and present case studies in indoor-outdoor scene classification and parts-based object detection. First, improved scene classification is accomplished using both low-level features, such as color and texture, and semantic features, such as the presence of sky and grass. Integration of low-level and semantic features is achieved using a Bayesian network framework. The network structure can be determined by expert opinion or by automated structure learning methods. Second, object detection at multiple views relies on a parts-based approach, where specialized detectors locate object parts and a Bayesian network acts as the arbitrator in order to determine the object presence. In general, Bayesian networks are found to be powerful integrators of different features and help improve the performance of image understanding systems.

Download Full-text

Analyzing the Effects of the Dwelling Characteristics on Space Heating Costs with the Rank Correlation Bayesian Network Model

10.21203/rs.3.rs-436208/v1 ◽

2021 ◽

Author(s):

Volkan Sevinç

Keyword(s):

Bayesian Networks ◽

Bayesian Network ◽

Network Model ◽

Building Materials ◽

Structure Learning ◽

Rank Correlation ◽

Space Heating ◽

Bayesian Network Model ◽

Heating Systems ◽

External Wall

Abstract Energy is one of the main concerns of humanity because energy resources are limited and costly. In order to reduce the costs and to use the energy for space heating effectively, new building materials, techniques and insulations facilities are being developed. Therefore, it is important to know which factors affect the space heating costs. This study aims to introduce the novel Rank Correlation Bayesian Network model and its application in analyzing the effects of dwelling characteristics on the space heating costs. The results show that the constructed Rank Correlation Bayesian Network model performed better than the Bayesian networks models estimated by Bayesian search, PC and Greedy Thick Thinning algorithms, which are kinds of structure learning algorithms having different kinds of estimation mechanisms to build Bayesian networks. The constructed Rank Correlation Bayesian Network model shows that the space heating costs of the dwellings are mostly affected by the heating systems used. Coal stoves, air conditioners and electric stoves appear to be the costliest heating systems. The second most important factor appears to be the existence of external wall insulation. The lack of external wall insulation almost doubles the space heating costs. The third most important factor is the building age. Dwellings on the ground floors and the first floors appear to pay the highest space heating costs. Therefore, dwellings on these floors need to be more effectively insulated. As the size of the dwelling increases the heating cost increases too. Another result is that facing directions and floor levels of the dwellings have the least effects on their space heating.

Download Full-text

Hybrid Optimization Algorithm for Bayesian Network Structure Learning

Information ◽

10.3390/info10100294 ◽

2019 ◽

Vol 10 (10) ◽

pp. 294 ◽

Cited By ~ 2

Author(s):

Xingping Sun ◽

Chang Chen ◽

Lu Wang ◽

Hongwei Kang ◽

Yong Shen ◽

...

Keyword(s):

Artificial Intelligence ◽

Bayesian Networks ◽

Bayesian Network ◽

Network Structure ◽

Structure Learning ◽

Expert Knowledge ◽

Hybrid Optimization ◽

Experimental Simulation ◽

Bayesian Network Structure ◽

Bayesian Network Structure Learning

Since the beginning of the 21st century, research on artificial intelligence has made great progress. Bayesian networks have gradually become one of the hotspots and important achievements in artificial intelligence research. Establishing an effective Bayesian network structure is the foundation and core of the learning and application of Bayesian networks. In Bayesian network structure learning, the traditional method of utilizing expert knowledge to construct the network structure is gradually replaced by the data learning structure method. However, as a result of the large amount of possible network structures, the search space is too large. The method of Bayesian network learning through training data usually has the problems of low precision or high complexity, which make the structure of learning differ greatly from that of reality, which has a great influence on the reasoning and practical application of Bayesian networks. In order to solve this problem, a hybrid optimization artificial bee colony algorithm is discretized and applied to structure learning. A hybrid optimization technique for the Bayesian network structure learning method is proposed. Experimental simulation results show that the proposed hybrid optimization structure learning algorithm has better structure and better convergence.

Download Full-text

Falcon Optimization Algorithm for Bayesian Networks Structure Learning

Computer Science ◽

10.7494/csci.2021.22.4.3773 ◽

2021 ◽

Vol 22 (4) ◽

Author(s):

Shahab Wahhab Kareem ◽

Mehmet Cudi Okur

Keyword(s):

Bayesian Networks ◽

Bayesian Network ◽

Optimization Algorithm ◽

Structure Learning ◽

Confusion Matrix ◽

Optimal Solution ◽

Score Function ◽

Scientific Models ◽

Data Sets ◽

Networks Structure

In machine-learning, one of the useful scientific models for producing the structure of knowledge is Bayesian network, which can draw probabilistic dependency relationships between variables. The score and search is a method used for learning the structure of a Bayesian network. The authors apply the Falcon Optimization Algorithm (FOA) as a new approach to learning the structure of Bayesian networks. This paper uses the Reversing, Deleting, Moving and Inserting operations to adopt the FOA for approaching the optimal solution of Bayesian network structure. Essentially, the falcon prey search strategy is used in the FOA algorithm. The result of the proposed technique is compared with Pigeon Inspired optimization, Greedy Search, and Simulated Annealing using the BDeu score function. The authors have also examined the performances of the confusion matrix of these techniques utilizing several benchmark data sets. As shown by the evaluations, the proposed method has more reliable performance than the other algorithms including producing better scores and accuracy values.

Download Full-text

Learning Discrete Bayesian Networks from Continuous Data

Journal of Artificial Intelligence Research ◽

10.1613/jair.5371 ◽

2017 ◽

Vol 59 ◽

pp. 103-132 ◽

Cited By ~ 9

Author(s):

Yi-Chun Chen ◽

Tim A. Wheeler ◽

Mykel J. Kochenderfer

Keyword(s):

Bayesian Networks ◽

Bayesian Network ◽

Structure Learning ◽

Minimum Description Length ◽

Real Data ◽

Continuous Variables ◽

Bayesian Network Structure ◽

Bayesian Network Structure Learning ◽

Accuracy Speed ◽

Standard Techniques

Learning Bayesian networks from raw data can help provide insights into the relationships between variables. While real data often contains a mixture of discrete and continuous-valued variables, many Bayesian network structure learning algorithms assume all random variables are discrete. Thus, continuous variables are often discretized when learning a Bayesian network. However, the choice of discretization policy has significant impact on the accuracy, speed, and interpretability of the resulting models. This paper introduces a principled Bayesian discretization method for continuous variables in Bayesian networks with quadratic complexity instead of the cubic complexity of other standard techniques. Empirical demonstrations show that the proposed method is superior to the established minimum description length algorithm. In addition, this paper shows how to incorporate existing methods into the structure learning process to discretize all continuous variables and simultaneously learn Bayesian network structures.

Download Full-text