Computational Intelligence and Neuroscience / 2015 / Article / Tab 1

Research Article

Learning Document Semantic Representation with Hybrid Deep Belief Network

Table 1

Document classification accuracy and document retrieval precision.

DatasetModelL_rDocument classification Document retrieval
Output units number Output units number
501001285121000501001285121000

20 NewsgroupsRSM0.0163.1166.8267.6766.5269.3162.8267.8068.6465.1067.48
0.00160.9462.9171.5573.6873.0556.1058.3269.1769.6569.34
0.000136.5962.7563.3370.4071.3828.4658.5659.3966.3765.14
DBN0.0172.7772.3572.4972.5672.3967.5267.7868.9069.2769.81
DocNADE0.0173.8373.4073.5474.7574.1168.5068.7870.1770.6670.82
HDBN0.0176.5776.6876.3576.4976.8976.9171.8972.9973.5073.91

BBC dataRSM0.0194.0494.6595.4195.1195.8793.4493.2992.9893.9094.81
0.00195.1196.4896.6494.6596.6494.6695.3496.1895.7996.03
0.000160.4094.6594.8096.7994.9570.8891.7694.0595.8895.42
DBN0.0195.3194.8196.0395.0695.0593.1894.1094.3094.1394.69
DocNADE0.0195.6096.7796.9396.8996.9394.9495.4496.2895.9896.13
HDBN0.0197.4097.4098.0197.0997.2595.4296.0296.9496.4896.94

We are committed to sharing findings related to COVID-19 as quickly as possible. We will be providing unlimited waivers of publication charges for accepted research articles as well as case reports and case series related to COVID-19. Review articles are excluded from this waiver policy. Sign up here as a reviewer to help fast-track new submissions.