Daricha Sutivong's research while affiliated with Chulalongkorn University and other places

Publications (21)

Article
Purpose Prediction markets are techniques to aggregate dispersed public opinions via market mechanisms to predict uncertain future events’ outcome. Many experiments have shown that prediction markets outperform other traditional forecasting methods in terms of accuracy. Logarithmic market scoring rules (LMSR) is one of the most simple and widely u...
Article
Full-text available
The aim of this paper is twofold: to propose the model of artificial prediction markets that capture the characteristics of real prediction markets and to study the impact of key parameters on the performance of the proposed markets. In the experiments, the artificial markets are implemented and the market performance in terms of convergence speed...
Conference Paper
Customer churn prediction plays a significant role in various businesses such as telecommunication, banking, and insurance. Accurate churn prediction helps businesses put forward a direct measure toward customers with tendency to leave rather than offering a generic promotion to all. Likewise, mobile application aims not only to acquire new custome...
Conference Paper
Pari-mutuel prediction markets are the markets where participants, based on their beliefs, buy contracts that have payoffs depending on a correct prediction of the future outcome. Implementing pari-mutuel prediction markets offers simplicity and liquidity guarantee. However, an appropriate trading duration is difficult to determine. While longer tr...
Conference Paper
Prediction markets are methodologies to aggregate dispersed knowledge on uncertainty from the crowd via a market mechanism. Various experiments using the prediction markets in real-world applications yield better accuracy than the polls. Most research on prediction markets focus on technical details of a particular market type. However, no guidelin...
Article
The real options technique has emerged as an evaluation tool for investment under uncertainty. It explicitly recognizes future decisions, and the exercise strategy is based on the optimal decisions in future periods. This paper employs the optimal stopping policy derived from real options approach to analyze and evaluate genetic algorithms, specifi...
Conference Paper
Feedforward neural networks are particularly useful in learning a training dataset without prior knowledge. However, weight adjusting with a gradient descent may result in the local minimum problem. Repeated training with random starting weights is among the popular methods to avoid this problem, but it requires extensive computational time. This p...
Article
This paper proposes an integrated model for water resource management using multi-objective optimization and rainfall forecast. To optimally allocate limited water supply, the objective functions involve maximizing net economic benefit while striving to maintain parity of water shortage in allocated areas. Moreover, because rain water is an importa...
Conference Paper
Full-text available
This paper proposes using a decision contour derived from real options analysis, which is an evaluation tool for investment under uncertainty, to suggest an optimal stopping time of the compact genetic algorithm on the trap problem. The proposed criterion provides a stopping boundary, where termination is optimal on one side and continuation is on...
Conference Paper
This paper proposes a quantitative model for balancing and optimizing portfolio of R&D projects. The model focuses on two dimensions of uncertainty - market and technical - to formulate R&D portfolio budget allocation problem. The investment is broken into two critical stages, namely R&D phase and commercialization phase. The real options analysis...
Conference Paper
The Bayesian networks support resource allocation in software project and also help in analyzing trade-offs among resources. The model predicts the probability distribution of every variable given incomplete data. Even though the Bayesian networks conveniently facilitate scenario-based analysis, they do not support finding an optimal solution in mu...
Conference Paper
Full-text available
The real options technique has emerged as an evaluation tool for investment under uncertainty. It explicitly recognizes future decisions, and the exercise strategy is based on the optimal decisions in future periods. The real options approach has been applied to many economic and financial problems, but few are in computer science and engineering....
Conference Paper
Traditional hardware design is not flexible. Specification changes often require redesigning from scratch. The objected-oriented (OO) concept, which consists of encapsulation, inheritance and reusability mechanisms, is applied to solve the problem. However, the OO concept still confronts code maintenance problem, as the same function may be redunda...
Conference Paper
Full-text available
The compact genetic algorithm (cGA) has a distinct characteristic that it requires almost minimal memory to store candidate solutions. It represents a population structure as a probability distribution over the set of solutions. Although cGA offers many advantages, it has a limitation that hinges on an assumption of the independency between each in...
Conference Paper
Software development projects are subject to external and internal risks that cause delays, budget overrun and poor quality. Portfolio management can be used to alleviate this problem, as it pools resources together and allows for resource sharing among projects. Consequently, projects are more likely to succeed. However, portfolio management using...
Conference Paper
Developing a complex system requires partitioning the target system into several subsystems. It is generally difficult to define each subsystem's scope and functional requirements as well as data dependencies among subsystems. Requirements Integration Model (RIM), which consists of a workflow model and a work procedure, can provide specific guideli...
Conference Paper
The resource decisions in software project using cost models do not satisfy managerial decision, as it does not support trade-off analysis among resources. A Bayesian net approach enables this analysis; however, it requires discretization of parameter values and thus sacrifices accuracy. Although narrow intervals can alleviate this problem, the num...
Conference Paper
Full-text available
In this paper we present a prediction process of the Stock Exchange of Thailand index using adaptive evolution strategies. The prediction process does not require the knowledge of the functional form a priori. In each recursion step, genetic algorithm is used to evolve the structure of the prediction function, whereas the coefficient is evolved by...

Citations

... . This is the softmax function (Boltzmann distribution with constant β = −k/T for fixed k and T ) of the inputs (q 0 t , q 1 t ). The β term is a liquidity factor [61] that adjusts the amount the price will increase or decrease given a change in the asset quantities. By using a Boltzmann distribution, the prices can be interpreted as probabilities. ...
... Babel et al. [16] developed a multi-objective water allocation model using the SICCON technique to support reservoir operators and managers in optimizing water allocation for a hypothetical reservoir. Using multi-objective programming and rainfall forecasts, Khummongkol et al. [21] developed a multi-objective integrated model for optimizing water allocation and management so as to maximize the NEB. By considering the water resources security, Wang et al. [22] developed a multi-objective water allocation model to improve the eco-environmental and socio-economic benefits in Zhangjiakou region of northern China. ...
... Prediction and Planning [1,2,5,9,16, Control and operational management [78][79][80][81][82][83][84][85][86][87][88] Process investigation [17,[89][90][91][92][93][94][95][96][97] Risk management [98][99][100][101][102][103][104][105][106][107] Process improvement [108][109][110][111][112][113][114] Technology adoption [8,115] Training and learning [116][117][118][119] SOUSA ET AL. ...
... Ha et al. [20] proposed the use of more than one probability vector (PV) to enhance the exploration properties of the algorithm. Rimcharoen et al. [21] improved the updating strategy of cGA by using a moving average technique (mcGA). Ahn and Ramakrishna [22] adopted 'elitism', i.e. the idea of reserving the best solution in each generation. ...
... Hewlett-Packard Corporation uses this technique to forecast sales [2]. Eli Lilly uses markets to predict the chance of new drugs passing the product test [3][4]. Google has run a number of prediction markets to forecast company performance and industry trends [5]. ...
... Complementarily to this, the RECSS method also uses a decomposition process through which one can build quality models for the system modules based on the ISO/IEC software quality standard. Other RE researchers suggested the use of process modelling tiers to manage the complexity of enterprise and ES process modeling [18], the technique of the Requirement Integration Model [32] to account for interdependencies in business workflows, and the Data Activity Model for Configuration approach [34] meant to help align a package to the organization by the joint engineering of data and process requirements. The authors of [3, 52], also proposed ontology-based approaches to the representation and gap analysis of enterprise requirements and packageembedded functionality. ...
... Also, our proposal is different from a framework because JPEAG can provide the programmers a entire runnable code. In terms of metric for measuring programming effort, COCOMO and COCOMO II have been used in many works such as [7], [8] and [9], for example. Furthermore, COCOMO II has been the bottom line for many researches whose aim is to improve the quality of the effort measurements such as [10][11] [12]. ...
... For the ANN classifier applied in this study, traditional feed-forward network with single hidden layer that includes 15 neurons was applied. The training process for each model is repeated more than three times to avoid the problem of convergence to a satisfactory solution [20]. Additionally, gradient descent with momentum and adaptive learning rate back-propagation (traingdx) has been used as implemented in the Matlab software. ...
... The studies [25][26] had shown how to define an optimal stopping time in genetic algorithms. Usually, an assigned maximum number of iterations is used to stop the generations, but how big the maximum number should be set is still unknown. ...
... The adaptive evolution strategies (A-ES) was first proposed in [12], into which a crossover technique has been added later by [2]. The A-ES is a combination of genetic algorithm (GA) and evolution strategies (ES). ...