基于特征组分层与半监督学习的鼠标轨迹识别

传统时间序列分类方法存在鼠标轨迹特征挖掘不充分、数据不平衡与标记样本量少等问题,造成识别效果较差。结合特征组分层和半监督学习,提出一种鼠标轨迹识别方法。通过不同视角构建有层次的鼠标轨迹特征组,并借鉴半监督学习的思想,利用多个随机森林模型对未标记样本进行伪标记,且将抽取标签预测一致且置信度较高的部分样本加入到训练集中。基于基础特征组和辅助特征组,在扩充后的训练集上训练随机森林模型,以实现鼠标轨迹的人机识别。实验结果表明,该方法可有效识别鼠标轨迹,且精确率、召回率与调和均值分别达到97.83%、94.72%和96.56%。     

基于LightGBM算法的能见度预测模型

为了提高能见度预报的准确率,尤其是低能见度预报的准确率,提出一种基于集成学习随机森林和LightGBM的能见度预测模型。首先,以数值模式系统的气象预报数据为基础,结合地面气象观测数据、PM_2.5浓度观测数据,利用随机森林算法构建特征向量;其次,针对不同时间跨度的缺失数据,设计了3种缺失值处理方法对缺失值进行替代,生成用于训练和测试的连续性较好的数据样本集;最后,建立基于LightGBM的能见度预测模型,并用网络搜索法对其进行参数优化。把所提模型与支持向量机（SVM）、多元线性回归（MLR）、人工神经网络（ANN）在性能上进行对比。实验结果表明,对于不同的等级的能见度,应用LightGBM的能见度预测模型获得预兆得分（TS）均较高,而对于<2 km的低能见度,该模型对各观测站点的能见度预测值与各观测站点的能见度实况值的平均相关系数为0.75,平均均方误差为6.49。可见基于LightGBM的预测模型能有效提高能见度预测精度。     

面向并发程序中锁机制的智能化推荐方法

软件开发人员在并发编程中面临着各种Java锁的选择。针对如何选择合适的锁机制进而提高程序性能的问题,提出了一种帮助并发程序开发人员选择锁机制的推荐方法LockRec。首先,采用程序静态分析技术,对并发程序中锁机制的使用进行分析并确定影响程序性能的程序特征属性;然后,使用改进的随机森林算法构建锁机制推荐模型,以帮助开发人员在同步锁、可重入锁、读写锁、邮戳锁这四种锁之中进行选择。分别选取了四种现有的机器学习数据集对LockRec进行了实验,所提方法的推荐准确率均值可达95.1%;此外使用真实并发程序对LockRec的推荐结果进行分析。实验结果表明LockRec可以有效提高并发程序的执行效率。     

基于增强多维多粒度级联森林的信用评分模型

信用风险是商业银行所面临的主要金融风险之一,而传统的基于统计学习的信用评分方法不能有效利用现有的特征学习方法,因此预测准确度不高。为解决这个问题,提出一种增强多维多粒度级联森林的方法建立信用评分模型,借鉴残差学习的思想,建立了多维多粒度级联残差森林（grcForest）模型,从而大幅增加提取的特征。除此之外,使用多维多粒度的扫描尽可能多地提取原始数据的特征,从而提高了特征提取的效率。对各模型的实验结果通过AUC（Area Under Curve）、准确率等指标进行评价,同时把所提模型与现有的统计学习和机器学习算法在四个不同的信用评分数据集上进行对比,可知所提出的模型的AUC值相较于轻量级梯度提升机（LightGBM）方法平均高1.13%,相较于极端梯度提升（XGBoost）方法平均高1.44%。从实验结果可以看出,提出的模型预测效果最佳。     

基于区域特征融合的RGBD显著目标检测

为了对各类自然场景中的显著目标进行检测,本文提出了一种将图像的深度信息引入区域显著性计算的方法,用于目标检测。首先对图像进行多尺度分割得到若干区域,然后对区域多类特征学习构建回归随机森林,采用监督学习的方法赋予每个区域特征显著值,最后采用最小二乘法对多尺度的显著值融合,得到最终的显著图。实验结果表明,本文算法能较准确地定位RGBD图像库中每幅图的显著目标。     

手机视频体验评估及建网方法研究

视频业务逐渐成为继语音业务之后的基础业务,本文基于U-vMOS评价体系,研究手机视频感知质量及体验评估,基于路测和SEQ数据,通过建模研究手机视频体验建网方法。     

Human Action Recognition Based on Dense Trajectories Analysis and Random Forest

This paper presents a human action recognition method. It analyzes the spatio-temporal grids along the dense trajectories and generates the histogram of oriented gradients (HOG) and histogram of optical flow (HOF) to describe the appearance and motion of the human object. Then, HOG combined with HOF is converted to bag-of-words (BoWs) by the vocabulary tree. Finally, it applies random forest to recognize the type of human action. In the experiments, KTH database and URADL database are tested for the performance evaluation. Comparing with the other approaches, we show that our approach has a better performance for the action videos with high inter-class and low inter-class variabilities.     

Explainable Artificial Intelligence Solution for Online Retail

Artificial intelligence (AI) and machine learning (ML) help in making predictions and businesses to make key decisions that are beneficial for them. In the case of the online shopping business, it’s very important to find trends in the data and get knowledge of features that helps drive the success of the business. In this research, a dataset of 12,330 records of customers has been analyzed who visited an online shopping website over a period of one year. The main objective of this research is to find features that are relevant in terms of correctly predicting the purchasing decisions made by visiting customers and build ML models which could make correct predictions on unseen data in the future. The permutation feature importance approach has been used to get the importance of features according to the output variable (Revenue). Five ML models i.e., decision tree (DT), random forest (RF), extra tree (ET) classifier, Neural networks (NN), and Logistic regression (LR) have been used to make predictions on the unseen data in the future. The performance of each model has been discussed in detail using performance measurement techniques such as accuracy score, precision, recall, F1 score, and ROC-AUC curve. RF model is the best model among all five chosen based on accuracy score of 90% and F1 score of 79% followed by extra tree classifier. Hence, our study indicates that RF model can be used by online retailing businesses for predicting consumer buying behaviour. Our research also reveals the importance of page value as a key feature for capturing online purchasing trends. This may give a clue to future businesses who can focus on this specific feature and can find key factors behind page value success which in turn will help the online shopping business.     

Potential of Unconventional Seed Oils and Fats from West African Trees: A Review of Fatty Acid Composition and Perspectives

Seed oils sourced from West Africa are generally not well-characterized, but likely to have an untapped potential. This review aims to make an overview of fatty acid (FA) composition of unconventional seed oils from semi-arid West African trees and evaluate potential for new and enhanced uses and for improving local livelihoods and biodiversity conservation. A total of 111 studies on FA composition were found, covering 31 species. Only 69 of the studies (62%) were included in the review, as 38% had unreliable or incomplete results. There was a clear link between taxonomic kinship and FA composition. Over 20 potentially interesting and underexploited oils were found, including oils with properties similar to palm oil, olive oil, coconut oil, shea butter, and cotton seed oil. About half of the oils have promising potential for cosmetics. One third of the oils were relatively saturated, indicating properties for structuring food and heat resistance. Most of the species had multiple uses and oil production could be profitable in co-production with other non-timber forest products. Furthermore, establishment of sustainable oil production and domestication of oil trees could promote biodiversity conservation. Enhanced oil production in semi-arid West Africa is promising, but several practical constraints remain to be overcome.