搜索结果: 1-6 共查到“Co-training”相关记录6条 . 查询时间(0.125 秒)
基于Co-training的用户属性预测研究
用户属性 Co-training 稀疏自编码器 梯度下降算法
2018/7/4
针对当前基于第三方应用数据的用户属性预测算法较少考虑应用前台实际使用时长问题，提出了“手机应用前台均使用时长”特征，同时采用基于稀疏自编码器和神经网络的Co-training框架，充分利用大量未标注数据，从应用类别和前台均使用时长两个角度进行属性预测。实验进行时，首先使用未标注数据对网络进行初始化，然后采用基于准确率的梯度下降算法对参数进行更新。实验结果表明，本文算法提高了用户属性预测准确率。
Bilingual Co-Training for Sentiment Classification of Chinese Product Reviews
Bilingual Co-Training Sentiment Classification Chinese Product
2015/9/9
The lack of reliable Chinese sentiment resources limits research progress on Chinese sentiment classification. However, there are many freely available English sentiment resources on the Web. This art...
Statistical Methods for Analyzing Tissue Microarray Images - Algorithmic Scoring and Co-training
Statistical Methods Analyzing Tissue Microarray Images Algorithmic Scoring Co-training
2011/3/18
Recent advances in tissue microarray technology have allowed immuno-histochemistry to become a powerful medium-to-high throughput analysis tool, particularly for the validation of diagnostic and progn...
Weakly-Supervised Violence Detection in Movies with Audio and Video Based Co-training
Violence Weakly-supervised pLSA Audio Video Co-training
2013/7/16
In this work, we present a novel method to detect violent shots in movies. The detection process is split into two views–––the audio and video views. From the audio-view, a weakly-supervised method is...
基于Co-Training的协同目标跟踪
目标跟踪 联合训练 半监督学习
2009/8/13
运动目标跟踪是计算机视觉的核心问题之一，广泛应用于诸多领域。该文提出一种基于Co-Training半监督学习框架的目标跟踪方法。该方法融合2种互相独立的特征信息来描述目标模型，采用Co-Training来协同更新模型，有效避免了现有方法的误差累积问题。实验结果证明，该方法在复杂场景下仍能实现稳定有效的跟踪。
Weakly-supervised Violence Detection in Movies with Audio and Video Based Co-training
Weakly-supervised Violence Detection Movies Audio Video Based Co-training
2010/12/20
In this work, we present a novel method to detect violent shots in movies. The detection process is split into two views–––the audio and video views. From the audio-view, a weakly-supervised method is...