A semi-supervised classification approach based on restricted Boltzmann machine for fMRI data

Liu, Ning; Yao, Li; Zhao, Xiaojie

doi:10.1109/bci48061.2020.9061617

Cited by 3 publications

(5 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The kappa index rates confirm the foregoing results. The models developed by Li et al, 48 Iscen et al, 53 Liu et al, 52 and Li et al 50 presented values of accuracy below 0.9 for all tested databases. The proposed model reached a kappa value equal to or greater than 0.9; in the other databases, its kappa index exceeded 0.9, indicating an excellent classification in the transductive phase with 4% labeled samples.…”

Section: Resultsmentioning

confidence: 82%

“…The other models exhibited an accuracy of less than 0.9 in all of the remaining scenarios. The models developed by Li et al, 48 Iscen et al, 53 Liu et al, 52 and Li et al 50 showed accuracy values below 0.9 for all databases, using 1% of the labeled samples. Table 5 summarizes the transductive phase results for the databases used, considering L with 2% labeled samples.…”

Section: Resultsmentioning

confidence: 90%

“…Liu et al 52 proposed a semi‐supervised approach combining unsupervised representation learning and supervised classification to improve the semi‐supervised classification of functional magnetic resonance imaging data. The unsupervised representation learning model was constructed using a restricted Boltzmann machine, and the feature parameters from the unsupervised model were provided as constraints in the semi‐supervised training of the classification model.…”

Section: Related Workmentioning

confidence: 99%

“…In addition, a comparison is made using SEEDED k‐means, which is not a classifier, but a clustering method; nevertheless, similar to the aforementioned models, it implements self‐labeling. We also compared the model with another models found in literature, such as: Ensemble, 49 Li et al, 48 Iscen et al, 53 Liu et al, 52 and Li et al 50 …”

Section: Benckmark Experimentsmentioning

confidence: 99%

“…In addition, a comparison is made using SEEDED k-means, which is not a classifier, but a clustering method; nevertheless, similar to the aforementioned models, it implements self-labeling. We also compared the model with another models found in literature, such as: Ensemble, 49 Li et al, 48 Iscen et al, 53 Liu et al, 52 and Li et al 50 All of these selected base classifiers are among the 10 most influential data mining algorithms. 73 The hyperparameters used in the classifiers are described and summarized in Table 3.…”

Section: Comparasion Models and Base Classifiermentioning

confidence: 99%

See 4 more Smart Citations

Deep semi‐supervised classification based in deep clustering and cross‐entropy

et al. 2021

View full text Add to dashboard Cite

Self‐labeled techniques, a semi‐supervised classification paradigm (SSC), are highly effective in alleviating the scarcity of labeled data used in classification tasks through an iterative process of self‐training. This problem was addressed by several approaches with different assumptions about the features of the input data, examples of these approaches being self‐training, co‐training, STRED, among others. This paper presents a framework for data self‐labeling based on deep autoencoder combined with a self‐labeled technique that takes into consideration cross‐entropy. The model uses the Encoder to reduce the dimensionality of the input that is submitted to a labeling layer. The weights of this layer are adjusted through the learning from a clustering performed in the Z space, which is the reduced dimensionality space. Results showed that the proposed method obtained competitive performance in relation to classic methods that are found in the literature.

show abstract