Yancong Zhou

Dr. Yancong Zhou

orcid

Research Focus & Metrics

Academic Career Length

20Years (2007-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 April 2013 in The International Journal of Advanced Manufacturing Technology
Verified
Journal Article
Closed access
PDF file
Sun XiaochenSun Xiaochen
Sun Xiaochen
YJ LiYJ Li
YJ Li
KGKannan Govindan
Kannan Govindan
YZYancong Zhou
Yancong Zhou

28 December 2022 in Computational Intelligence and Neuroscience
Verified
Journal Article
Open access
PDF file
YZYancong Zhou
Yancong Zhou
QZQian Zhang
Qian Zhang
DWDongdong Wang
Dongdong Wang
XGXiaoying Gu
Xiaoying Gu
The research of text sentiment analysis based on deep learning is increasingly rich, but the current models still have different degrees of deviation in understanding of semantic information. In order to reduce the loss of semantic information and improve the prediction accuracy as much as possible, the paper creatively combines the doc2vec model with the deep learning model and attention mechanism and proposes a new hybrid sentiment analysis model based on the doc2vec + CNN + BiLSTM + Attention. The new hybrid model effectively exploits the structural features of each part. In the model, the understanding of the overall semantic information of the sentence is enhanced through the paragraph vector pretrained by the doc2vec structure which can effectively reduce the loss of semantic information. The local features of the text are extracted through the CNN structure. The context information interaction is completed through the bidirectional cycle structure of the BiLSTM. The performance is improved by allocating weight and resources to the text information of different importance through the attention mechanism. The new model was built based on Keras framework, and performance comparison experiments and analysis were performed on the IMDB dataset and the DailyDialog dataset. The results have shown that the accuracy of the new model on the two datasets is 91.3% and 93.3%, respectively, and the loss rate is 22.1% and 19.9%, respectively. The accuracy on the IMDB datasets is 1.0% and 0.5% higher than that of the CNN-BiLSTM-Attention model and ATT-MCNN-BGRUM model in the references. Comprehensive comparison has shown the overall performance is improved, and the new model is effective.

11 October 2023 in PLOS ONE
Verified
Journal Article
Open access
PDF file
YZYancong Zhou
Yancong Zhou
WCWenyue Chen
Wenyue Chen
Sun XiaochenSun Xiaochen
Sun Xiaochen
DYDandan Yang
Dandan Yang
Analyzing customers’ characteristics and giving the early warning of customer churn based on machine learning algorithms, can help enterprises provide targeted marketing strategies and personalized services, and save a lot of operating costs. Data cleaning, oversampling, data standardization and other preprocessing operations are done on 900,000 telecom customer personal characteristics and historical behavior data set based on Python language. Appropriate model parameters were selected to build BPNN (Back Propagation Neural Network). Random Forest (RF) and Adaboost, the two classic ensemble learning models were introduced, and the Adaboost dual-ensemble learning model with RF as the base learner was put forward. The four models and the other four classical machine learning models-decision tree, naive Bayes, K-Nearest Neighbor (KNN), Support Vector Machine (SVM) were utilized respectively to analyze the customer churn data. The results show that the four models have better performance in terms of recall rate, precision rate, F1 score and other indicators, and the RF-Adaboost dual-ensemble model has the best performance. Among them, the recall rates of BPNN, RF, Adaboost and RF-Adaboost dual-ensemble model on positive samples are respectively 79%, 90%, 89%,93%, the precision rates are 97%, 99%, 98%, 99%, and the F1 scores are 87%, 95%, 94%, 96%. The RF-Adaboost dual-ensemble model has the best performance, and the three indicators are 10%, 1%, and 6% higher than the reference. The prediction results of customer churn provide strong data support for telecom companies to adopt appropriate retention strategies for pre-churn customers and reduce customer churn.

21 March 2018 in Discrete Dynamics in Nature and Society
Verified
Journal Article
Open access
PDF file
YZYancong Zhou
Yancong Zhou
JFJin Feng
Jin Feng
JWJie Wei
Jie Wei
Sun XiaochenSun Xiaochen
Sun Xiaochen
Supply disruption may cause strong complaints of customers, which is a cost loss for the firms in the supply chain. Obviously, if realizing that there is the disruption risk, the members in a supply chain will adjust their decisions. For analyzing the influence, we consider a popular supply chain mode with dual channels, where one manufacturer has its direct sales channel and one traditional retailer channel. The manufacturer may suffer a supply disruption so that all ordered products by the retailer or the direct retail channel will be lost, and the members in supply chain will bear the corresponding disruption penalty from the customers. By considering four structures with different market power relations, the closed-form optimal price decisions of the four models are given. We found that the disruption factor improves the sales prices for any member structure as compared to the supply chain without the disruption. And the direct retail prices in the different modes are the same as each other, but the price of the traditional channel is influenced by the market share. And the sorts of the sales prices under different structures are given. We also conduct some extensive numerical analysis and compare the results under different structures. We observe that the expected optimal profits of considering the external penalty are smaller than those of no external penalty, and we give a sort of the optimal expected profits. And we also provide the effects of some parameters on the optimal decisions and the optimal expected profits.

14 July 2014 in Sustainability
Verified
Journal Article
Open access
PDF file
XHXiaonan Han
Xiaonan Han
Sun XiaochenSun Xiaochen
Sun Xiaochen
YZYancong Zhou
Yancong Zhou
This article studies a supply chain composed of a manufacturer and two competing retailers. The manufacturer produces two substitutable products and offers respective service levels to customers who buy one of the two products. Each retailer can only order one kind of product from the manufacturer, and then sell them to the market at a certain sale price. The demands for two products are influenced not only by the service levels the manufacturer provides, but also the sales prices of the two products. Furthermore, we investigate the equilibrium behavior of members in the supply chain with the aid of the Stackelberg game, and discover a number of insights concerning some important parameters. Finally, Numerical analysis is presented to validate our theoretical results and compare channel performances.

16 June 2014 in Discrete Dynamics in Nature and Society
Verified
Journal Article
Open access
PDF file
YZYancong Zhou
Yancong Zhou
YDYong-Feng Dong
Yong-Feng Dong
HXHong-Mei Xia
Hong-Mei Xia
JGJun-Hua Gu
Jun-Hua Gu
Routing optimization is a key technology in the intelligent warehouse logistics. In order to get an optimal route for warehouse intelligent vehicle, routing optimization in complex global dynamic environment is studied. A new evolutionary ant colony algorithm based on RFID and knowledge-refinement is proposed. The new algorithm gets environmental information timely through the RFID technology and updates the environment map at the same time. It adopts elite ant kept, fallback, and pheromones limitation adjustment strategy. The current optimal route in population space is optimized based on experiential knowledge. The experimental results show that the new algorithm has higher convergence speed and can jump out the U-type or V-type obstacle traps easily. It can also find the global optimal route or approximate optimal one with higher probability in the complex dynamic environment. The new algorithm is proved feasible and effective by simulation results.