Facial Expression Classification Using Artificial Neural Network and K-Nearest Neighbor

Hai, Tran Son; Thai, Le Hoang; Thuy, Nguyen Thanh

doi:10.5815/ijitcs.2015.03.04

Cited by 13 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, MATLAB R2019a is used for implementation [18]. In order to achieve more reliable results, all experiments have been reported by 10-fold cross validation [19]. In this paper, the real data of a number of subscribers Copyright © 2021 MECS I.J.…”

Section: Resultsmentioning

confidence: 99%

Modelling Electricity Consumption Forecasting Using the Markov Process and Hybrid Features Selection

Dalkani¹,

Mojarad²,

Arfaeinia³

2021

IJISA

View full text Add to dashboard Cite

Given the problem of electrical energy storage, it is critical to predict the amount of load required in order to have a reliable and stable power distribution network. Predicting electricity consumption of subscribers and analyzing their consumption behavior under the influence of various factors and time variables is important. Given the large volume of subscriber consumption data and the effective factors, it is only possible to analyze the data using new information technology tools such as data mining. In this paper, feature selection, clustering and Markov process techniques are used to model and predict the power consumption data of subscribers. First, the selection of a subset of effective features is based on the combined PCA approach and the Firefly algorithm. Subscribers are then clustered based on the features selected by the K-means. Finally, subscriber behavior patterns are modeled to predict consumption using the Markov process on high-risk clusters. This study is simulated based on the data of electricity subscribers in Bushehr-Iran Power Distribution Company. The simulation results show the superiority of the proposed model over other similar algorithms such as LASSO-QRNN and HyFIS. The accuracy of power consumption prediction in the proposed method is about 1% compared to LASSO-QRNN and about 0.5% compared to HyFIS.

show abstract

Section: Resultsmentioning

confidence: 99%

Modelling Electricity Consumption Forecasting Using the Markov Process and Hybrid Features Selection

Dalkani¹,

Mojarad²,

Arfaeinia³

2021

IJISA

View full text Add to dashboard Cite

show abstract

“…Then, the largest category of K samples is considered to be the label of the test image. The second method is the Artificial Neural Networks (ANN) [20] , which achieve image classification through a neural network. The third method is the Hidden Markov Model (HMM) [21] , by constructing an category distribution to calculate the probability of which category the test image belongs to.…”

Section: Related Workmentioning

confidence: 99%

Weakly supervised facial expression recognition via transferred DAL-CNN and active incremental learning

Liu

Zhai

et al. 2019

Soft Comput

View full text Add to dashboard Cite

In recent years, facial expression recognition (FER) has becoming a growing topic in computer vision with promising applications on virtual reality and human-robot interaction. Due to the influence of illumination, individual differences, attitude variation and etc., facial expression recognition with robust accuracy in complex environment is still an unsolved problem. Meanwhile, with the widely use of social communication, massive data is uploaded to the Internet, the effective utilization of those data is still a challenge due to noisy label phenomenon in the study of FER. To resolve the above-mentioned problems, Firstly, a double active layer (DAL) based CNN is established to recognize the facial expression with high accuracy by learning robust and discriminative features from the data, which could enhance the robustness of network. Secondly, an active incremental learning method was utilized to tackle the problem of using Internet data. During the training phase, a two-stage transfer learning method is explored to transfer the relative information from face recognition to FER task to alleviate the inadequate training data in deep convolution network. Besides, in order to make better use of facial expression data from website and further improve the FER accuracy, UFEDW (Unconstrained Facial Expression Database from Website) database is built in this paper. Extensive experiments performed on two public facial expression recognition database FER 2013 and SFEW 2.0 have demonstrated that the proposed scheme outperforms the state-of-the-art methods, which could achieve 67.08% and 51.90% respectively.

show abstract

“…In the recent decades, new methods such as artificial neural networks (ANNs) and adaptive neuro-fuzzy inference system (ANFIS) have been used to develop the Determined by MTT Assay on Cancer Cell Lines [12][13][14][15]. In this paper, the applicability of ANFIS and radial basis function (RBF) network, which is a type of ANNs, for prediction of IC50 values evaluated by MTT assay in human cancer cell lines are investigated.…”

Section: Introductionmentioning

confidence: 99%

The Use of ANFIS and RBF to Model and Predict the Inhibitory Concentration Values Determined by MTT Assay on Cancer Cell Lines

Rezaei¹,

Noori²,

Taghipour³

2016

IJITCS

View full text Add to dashboard Cite

Abstract-The computational intelligence such as artificial neural network (ANN) and fuzzy inference system (FIS) is a strong tool for prediction and simulation in engineering applications. In this paper, radial basis function (RBF) network and adaptive neuro-fuzzy inference system (ANFIS) are used for prediction of IC50 (the 50% inhibitory concentration) values evaluated by the MTT assay in human cancer cell lines. For developing of the proposed models, the input parameters are the concentration of the drug and the types of cell lines and the output is IC50 values in the A549, H157, H460 and H1975 cell lines. The predicted IC50 values using the proposed RBF and ANFIS models are compared with the experimental data. The obtained results show that both RBF and ANFIS models have achieved good agreement with the experimental data. Therefore, the proposed RBF and ANFIS models are useful, reliable, fast and cheap tools to predict the IC50 values determined by the MTT assay in human cancer cell lines.

show abstract

Facial Expression Classification Using Artificial Neural Network and K-Nearest Neighbor

Cited by 13 publications

References 14 publications

Modelling Electricity Consumption Forecasting Using the Markov Process and Hybrid Features Selection

Modelling Electricity Consumption Forecasting Using the Markov Process and Hybrid Features Selection

Weakly supervised facial expression recognition via transferred DAL-CNN and active incremental learning

The Use of ANFIS and RBF to Model and Predict the Inhibitory Concentration Values Determined by MTT Assay on Cancer Cell Lines

Contact Info

Product

Resources

About