Search

In order to solve deviation and imbalance of the traditional multi-class classification. This paper designs an improved localized multi-class classification algorithm based on mutual communication entropy and Support Vector Data Description (SVDD), know as EL-SVDD algorithm. First, this algorithm calculates parameter values of the mutual communication entropy with many local classes of samples. Second, one class is placed inside the multidimensional sphere based on the mutual communication entropy. Finally, according to the samples and parameter values of the mutual communication entropy, it reinterpreted the C values of SVDD algorithm. As the result, the experiments shows that EL-SVDD algorithm not only has the feasibility, but also can improve the accuracy analysis of multi-class classification stably and effectively.

Recommendation system is a commercial marketing method. What more, the system could increase adhesion and satisfaction of consumers to the website which brings great commercial benefit to electronic commerce. But with big data ages coming, it makes a great challenge to real-time recommendation system. As for latent factor class collaborative filtering algorithm, a distributed constructed latent factor algorithm based on cloud is presented in this paper. The algorithm could keep collaborative filtering in good recommendation and ensure the real time in massive data environment. The simulation shows that the algorithm could achieve the recommendation efficiently and quickly. High speedup and scalability are proved.

In the medical system, medical record is summarized, analyzed and predicted a patient seizure type and incidence. With the development of information technology, a large amount of data is to be processed. Traditional analysis algorithm could not be effectively processed to obtain the best predictive results as the data increasing. Decision tree algorithm based on cloud platform is used to record, analyze and predict patients medical data in this paper. A large number of experimental results show that distributed decision tree algorithm proposed in this paper is efficient and could complete prediction work in medical system. The algorithm has good expansibility, its very suitable for large-scale and multitude medical data process.

Personalized recommendation provides convenience to users and brings more benefit to companies as well. It has been an important part of electronic commerce website. Collaborative filtering is a common algorithm in recommendation system. But with massive data ages coming, traditional collaborative filtering algorithm could not finish recommendation in time. A neighbor model algorithm based on MapReduce distributed computing framework is presented against to collaborative filtering algorithm. The presented algorithm could accomplish the personalized recommendation effectively and meet the real time requirement completely. The simulation shows that the algorithm has high efficiency and could complete the recommended in a highly efficient and real-time.

To improve the validity and efficiency of feature recognition for 3D reconstruction from engineering drawings, this paper presents a new method of view loops separation based on multi-granularity information acquisition. By analyzing the line-frames of combined primitive views and hidden semantics in engineering drawing, combined relationship of graphics primitives can be identified, then applies line-frames partition method of figuration analysis, view loop separation of individual primitive is realized by inertial loop search with pre-set priority. This algorithm is actualized in AutoCAD 2008 platform with development tools of Object ARX2008 and Visual C#2.0. With this algorithm, an application module of 3D intelligent reconstruction system is developed and finally the validity of this research is verified by examples.

For inverse constructing object from its orthographic image, a new method of 3D reconstruction based on image vectorization and bi-arc approximation of cubic B-spline is investigated. A 3D CAD model is automatically acquired from 2D views by means of 2D primitive recognition, 3D reconstruction of primitive and combination of 3D primitive bodies. With the help of CimatronE8.0 software, reasonable procedure design and optimization toolpath planning are applied to increase machining accuracy and productivity during automatic programming for reconstructed 3D solid. Reconstructed parts are machined on the VMC-600 type machine center with FANUC system. The results of application indicate that this method has many advantages, such as quickly obtaining the CAD model of part, getting smooth machined surface, efficient machining without invalid or reduplicate feed and so on. As a result, it can be applied widely in NC machining field.