Chi-Hwa Song

Extended simulated annealing for augmented TSP and multi-salesmen TSP

An extended simulated annealing (ESA), based on grand canonical ensemble (GCE), is proposed. An ESA is used to solve the augmented traveling salesman problems (ATSP) and the multiple traveling salesmen problems. Experimental results show that ESA has salient features such as simplicity and ability to find high-quality solutions as simulated annealing has.

Extended Mean Field Annealing for Clustering Incomplete Data

Clustering is an very important research topic in knowledge discovery and machine learning. But sometimes the data set for clustering contains vectors missing one or more of the feature values, and is called as incomplete. The incomplete data problem exists in a wide range of field such as computer vision, biological system, and remote sensing. The problem of clustering of incomplete data is considered in this paper. In this paper, we propose a new extended MFA (mean field annealing) algorithm to solve the problem of clustering of incomplete data in the continuous-value state space and show the result of the experiment. The traditional fuzzy clustering methods calculate the centroid vectors of the clusters and then determined the membership probability, and repeat this process until the optimum solution is found. By contrast, the method proposed in this paper perturbs the membership probability, and determines whether to accept the perturbed state or not according to the changes of the energy. The result is compared with the optimal completion strategy fuzzy c-means (FCM) clustering of incomplete data algorithm and shows that the proposed method solves the problem of clustering incomplete data very well and gets a much better result.

A New Classifier to Deal with Incomplete Data

Classification is a very important research topic in knowledge discovery and machine learning. Decision-tree is one of the well-known data mining methods that are used in classification problems. But sometimes the data set for classification contains vectors missing one or more of the feature values, and is called as incomplete data. Generally, the existence of incomplete data will degrade the learning quality of classification models. If the incomplete data can be dealt well, the classifier can be used to real life applications. So handling incomplete data is important and necessary for building a high quality classification model. In this paper a new decision tree is proposed to solve the incomplete data classification problem and it has a very good performance. At the same time, the new method solves two other important problems: rule refinement problem and importance preference problem, which ensures the outstanding advantages of the proposed approach. Significantly, this is the first classifier which can deal with all these problems at the same time.

Applying UChooBoost algorithm in precision agriculture

Informatization is primary characteristic of current agriculture stage. Precision Agriculture is destination of new technologies and principles in agriculture using digital information. We consider data analysis as an aspect of Precision Agriculture and introduce UChooBoost applied to classification problem in agriculture. UChooBoost is a supervised learning ensemble-based algorithm for extended data, based on bootstrap technique. UChoo classifier is used as Weak Learner. Combining hypotheses by new weighted majority voting developed for extended results expression allows UChooBoost to achieve better performance level.

FLDF Based Decision Tree using Extended Data Expression

We introduce a classification algorithm which can be applied to a problem with a data set included a missing variable. In this algorithm we use data expansion treating it with a weight value and the probability techniques. It is applied to extending a classifier which is considered the optimal projection plane based on Fishers formula. For doing this, we derive equations from the procedure to be applied to the data expansion. The result is compared to that of different measurements by choosing one variable in the data set and then modifying the rate of missing and non-missing values in this selected variable. The result of a data set with non-missing variable compares with that of C4.5 which is known as a knowledge acquisition tool in machine learning.

A Mean Field Annealing Algorithm for Fuzzy Clustering

In the classical clustering, an item must entirely belong to a cluster. Fuzzy clustering, however, describes more accurately the ambiguous type of structure in data. Fuzzy clustering is useful for partitioning a set of objects into a certain number of groups by assigning the membership probabilities to each object. In fuzzy clustering, the membership of each datum in each cluster is represented by the membership matrix. In the proposed method, the elements of membership matrix are updated in parallel until they reach one of the global optimal solutions. It differs from the traditional fuzzy clustering methods. In classical fuzzy clustering, the centroid vectors of the clusters in the space are calculated, and then the membership probability matrix is determined, and the process is repeated until the optimum solution is found. By contrast, the method proposed here perturbs the membership probability, and determines whether the the perturbed state should be accepted or not according to the changes of the energy. One Variable Stochastic Simulated Annealing(OVSSA), a continuous valued version of the Mean Field Annealing(MFA) algorithm which is known as a massively parallel algorithm, is employed as an optimization technique. The MFA combines characteristics of the simulated annealing and the neural network and exhibits the rapid convergence of the neural network while preserving the solution quality afforded by Stochastic Simulated Annealing(SSA).

Building a Packages Delivery Schedule Using Extended Simulated Annealing

In this paper, we describe a package delivery problem and propose a method based on the extended simulated annealing(ESA) algorithm which is able to find an optimal routing path for efficient package delivery service. The problem domain that we solve is modeled as a weighted, directed graph.

Solving multi-sensor problem with a new approach

Smart environments is a technological concept that, according to Mark Weiser is ldquoa physical world that is richly and invisibly interwoven with sensors, actuators, displays, and computational elements, embedded seamlessly in the everyday objects of our lives, and connected through a continuous network.rdquo But sometimes the data gathered from the sensors is with different importance. It means some sensors are more reliable than others for some reasons. For example some sensors may be in relatively bad environments and some of the gathered data is destroyed or ruined. How to deal with the information gathered from different sensors efficiently is an important multi-sensors problem. The existence of multi-sensors problem will degrade the learning quality of classification models. And almost all of the existing classifier can not deal with this problem. So handling multi-sensors problem is important and necessary for building a high quality classification model and smart environments. In this paper a new classifier capable of dealing with this multi-sensors problem is proposed and it has a very good performance which is proved by experiments. This classifier can combine the information gathered from different sensors efficiently and in can add the new coming data to make a more efficient classifier even all of the original data is lost. Because of all the advantages it has, the new classifier is proposed sincerely to apply into smart environments.

A new classifier for numerical incomplete data

Classification of the numerical data is a very important research topic in machine learning. But the incomplete data is very common in real world application. And the existence of incomplete data degrades the learning quality of classification models. But the existence of incomplete data always decrease the quality of classification models, To show the definition of missing data more intuitively, The example is taken like this: If Xl=(l,2,3,4), then (?,2,3,4) is X with 25% incomplete data, and (1,?,?,4) is XI with 50% incomplete data. In this paper a new classifier is proposed to solve the incomplete data classification problem and it has an outstanding performance.

A Classifier Capable of Handling New Attributes

During knowledge acquisition, a new attribute can be added at any time. In such a case, rule generated by the training data with the former attribute set can not be used. Moreover, the rule can not be combined with the new data set with the newly added attribute(s) using the existing algorithms. In this paper, we propose further development of the new inference engine, UChoo, that can handle the above case naturally. Rule generated from the former data set can be combined with the new data set to form the refined rule. This paper shows how this can be done consistently by the extended data expression, and also shows the experimental result to claim the effectiveness of the algorithm