Janyl Jumadinova | University of Nebraska at Omaha (original) (raw)

Uploads

Papers by Janyl Jumadinova

Research paper thumbnail of Multi-attribute Regret-Based Dynamic Pricing

In this paper, we consider the problem of dynamic pricing by a set of competing sellers in an inf... more In this paper, we consider the problem of dynamic pricing by a set of competing sellers in an information economy where buyers differentiate products along multiple attributes, and buyer preferences can change temporally. Previous research in this area has either focused on dynamic pricing along a limited number of (e.g. binary) attributes, or, assumes that each seller has access to private information such as preference distribution of buyers, and profit/price information of other sellers. However, in real information markets, private information about buyers and sellers cannot be assumed to be available a priori. Moreover, due to the competition between sellers, each seller faces a tradeoff between accuracy and rapidity of the pricing mechanism. In this paper, we describe a multi-attribute dynamic pricing algorithm based on minimax regret that can be used by a seller’s agent called a pricebot, to maximize the seller’s utility. Our simulation results show that the minimax regret based dynamic pricing algorithm performs significantly better than other algorithms for rapidly and dynamically tracking consumer attributes without using any private information from either buyers or sellers.

Research paper thumbnail of The Pseudo Gradient Search and a Penalty Technique Used in Classiflcations

The aim of this work is to use the pseudo gradient search to solve classification problems. In mo... more The aim of this work is to use the pseudo gradient search to solve classification problems. In most classifiers, the goal is to reduce the misclassified rate that is discrete. Since pseudo gradient search is a local search, to use it for classification problem, objective function has to be real valued. A penalty technique is used for this purpose.

Research paper thumbnail of Firefly-Inspired Synchronization for Improved Dynamic Pricing in Online Markets

We consider the problem of dynamic pricing by sellers in an online market economy using software ... more We consider the problem of dynamic pricing by sellers in an online market economy using software agents called price bots. In previous research on dynamic pricing algorithms, each seller's pricebot employs either heuristics-based or learning-based techniques to determine and update the profit maximizing price for itself at certain intervals in response to changes in market dynamics. In these dynamic pricing techniques, each seller's pricebot uses only its private information such as past prices and profits to update its price in successive intervals. In this paper, we posit that the profits obtained by a pricebot can be improved if each pricebot incorporates its competitors' pricing information along with its private price and profit information in its price-update calculations. However, incorporating competitors' pricing information accurately into a pricebot's dynamic pricing algorithm is a challenging problem because competing sellers (pricebots) update their prices asynchronously and by an amount determined by each seller's private pricing strategy. Our contribution in this paper is a novel dynamic pricing algorithm that uses a distributed synchronization model observed in nature to align each seller's price with its competitors' prices. Our analytical and simulation results show that the combination of a heuristics-based pricing mechanism that uses only a seller's private information and the synchronization-based mechanism that aligns its prices with its competitors, enables a seller's pricebot to improve its profits by as much as 78% as compared to previous dynamic pricing algorithms.

Research paper thumbnail of A multi-agent system for analyzing the effect of information on prediction markets

International Journal of Intelligent Systems, 2011

Prediction markets have been shown to be a useful tool for forecasting the outcome of future even... more Prediction markets have been shown to be a useful tool for forecasting the outcome of future events by aggregating public opinion about the event's outcome. In this paper, we investigate an important aspect of prediction markets—the effect of different information-related parameters on the behavior of the traders in the market. We have developed a multi-agent based system that incorporates different information-related aspects including the arrival rate of information, the reliability of information, the penetration or accessibility of information among the different traders, and the perception or impact of information by the traders. We have performed extensive simulations of our agent-based prediction market for analyzing the effect of information-related parameters on the traders' behaviors expressed through their trading prices, and compared our agents' strategies with another agent-based pricing strategy used in prediction markets called the zero intelligence strategy. Our results show that information-related parameters have a significant impact on traders' beliefs about event outcomes, and, frequent, reliable information about events improves the utilities that the traders receive. Overall, our work provides a better understanding of the effect of information on the operation of prediction markets and on the strategies used by the traders in the market. © 2011 Wiley Periodicals, Inc.

Research paper thumbnail of A Multi-agent Prediction Market Based on Boolean Network Evolution

Prediction markets have been shown to be a useful tool in forecasting the outcome of future event... more Prediction markets have been shown to be a useful tool in forecasting the outcome of future events by aggregating public opinion about the events' outcome. Previous research on prediction markets has mostly analyzed the prediction markets by building complex analytical models. In this paper, we posit that simpler yet powerful Boolean rules can be used to adequately describe the operations of a prediction market. We have used a multi-agent based prediction market where Boolean network based rules are used to capture the evolution of the beliefs of the market's participants, as well as to aggregate the prices in the market. We show that despite the simplification of the traders' beliefs in the prediction market into Boolean states, the aggregated market price calculated using our BN model is strongly correlated with the price calculated by a commonly used aggregation strategy in existing prediction markets called the Logarithmic Market Scoring Rule (LMSR). We also empirically show that our Boolean network-based prediction market can stabilize market prices under the presence of untruthful belief revelation by the traders.

Research paper thumbnail of Multi-attribute Regret-Based Dynamic Pricing

In this paper, we consider the problem of dynamic pricing by a set of competing sellers in an inf... more In this paper, we consider the problem of dynamic pricing by a set of competing sellers in an information economy where buyers differentiate products along multiple attributes, and buyer preferences can change temporally. Previous research in this area has either focused on dynamic pricing along a limited number of (e.g. binary) attributes, or, assumes that each seller has access to private information such as preference distribution of buyers, and profit/price information of other sellers. However, in real information markets, private information about buyers and sellers cannot be assumed to be available a priori. Moreover, due to the competition between sellers, each seller faces a tradeoff between accuracy and rapidity of the pricing mechanism. In this paper, we describe a multi-attribute dynamic pricing algorithm based on minimax regret that can be used by a seller’s agent called a pricebot, to maximize the seller’s utility. Our simulation results show that the minimax regret based dynamic pricing algorithm performs significantly better than other algorithms for rapidly and dynamically tracking consumer attributes without using any private information from either buyers or sellers.

Research paper thumbnail of The Pseudo Gradient Search and a Penalty Technique Used in Classiflcations

The aim of this work is to use the pseudo gradient search to solve classification problems. In mo... more The aim of this work is to use the pseudo gradient search to solve classification problems. In most classifiers, the goal is to reduce the misclassified rate that is discrete. Since pseudo gradient search is a local search, to use it for classification problem, objective function has to be real valued. A penalty technique is used for this purpose.

Research paper thumbnail of Firefly-Inspired Synchronization for Improved Dynamic Pricing in Online Markets

We consider the problem of dynamic pricing by sellers in an online market economy using software ... more We consider the problem of dynamic pricing by sellers in an online market economy using software agents called price bots. In previous research on dynamic pricing algorithms, each seller's pricebot employs either heuristics-based or learning-based techniques to determine and update the profit maximizing price for itself at certain intervals in response to changes in market dynamics. In these dynamic pricing techniques, each seller's pricebot uses only its private information such as past prices and profits to update its price in successive intervals. In this paper, we posit that the profits obtained by a pricebot can be improved if each pricebot incorporates its competitors' pricing information along with its private price and profit information in its price-update calculations. However, incorporating competitors' pricing information accurately into a pricebot's dynamic pricing algorithm is a challenging problem because competing sellers (pricebots) update their prices asynchronously and by an amount determined by each seller's private pricing strategy. Our contribution in this paper is a novel dynamic pricing algorithm that uses a distributed synchronization model observed in nature to align each seller's price with its competitors' prices. Our analytical and simulation results show that the combination of a heuristics-based pricing mechanism that uses only a seller's private information and the synchronization-based mechanism that aligns its prices with its competitors, enables a seller's pricebot to improve its profits by as much as 78% as compared to previous dynamic pricing algorithms.

Research paper thumbnail of A multi-agent system for analyzing the effect of information on prediction markets

International Journal of Intelligent Systems, 2011

Prediction markets have been shown to be a useful tool for forecasting the outcome of future even... more Prediction markets have been shown to be a useful tool for forecasting the outcome of future events by aggregating public opinion about the event's outcome. In this paper, we investigate an important aspect of prediction markets—the effect of different information-related parameters on the behavior of the traders in the market. We have developed a multi-agent based system that incorporates different information-related aspects including the arrival rate of information, the reliability of information, the penetration or accessibility of information among the different traders, and the perception or impact of information by the traders. We have performed extensive simulations of our agent-based prediction market for analyzing the effect of information-related parameters on the traders' behaviors expressed through their trading prices, and compared our agents' strategies with another agent-based pricing strategy used in prediction markets called the zero intelligence strategy. Our results show that information-related parameters have a significant impact on traders' beliefs about event outcomes, and, frequent, reliable information about events improves the utilities that the traders receive. Overall, our work provides a better understanding of the effect of information on the operation of prediction markets and on the strategies used by the traders in the market. © 2011 Wiley Periodicals, Inc.

Research paper thumbnail of A Multi-agent Prediction Market Based on Boolean Network Evolution

Prediction markets have been shown to be a useful tool in forecasting the outcome of future event... more Prediction markets have been shown to be a useful tool in forecasting the outcome of future events by aggregating public opinion about the events' outcome. Previous research on prediction markets has mostly analyzed the prediction markets by building complex analytical models. In this paper, we posit that simpler yet powerful Boolean rules can be used to adequately describe the operations of a prediction market. We have used a multi-agent based prediction market where Boolean network based rules are used to capture the evolution of the beliefs of the market's participants, as well as to aggregate the prices in the market. We show that despite the simplification of the traders' beliefs in the prediction market into Boolean states, the aggregated market price calculated using our BN model is strongly correlated with the price calculated by a commonly used aggregation strategy in existing prediction markets called the Logarithmic Market Scoring Rule (LMSR). We also empirically show that our Boolean network-based prediction market can stabilize market prices under the presence of untruthful belief revelation by the traders.