| A Web Service for Accommodation Renting: Evolutionary Bargaining Approach (2007) | |||||||||||||
Abstract | |||||||||||||
| This project developed a web service to search for and bargain on, on behalf of students, good accommodation in the dynamic market that is accommodation letting, where students and landlords could come and go at any time. The landlords registered their accommodation information through a website. Then on behalf of students the web service can autonomously watch the market's changing and use evolutionary searching techniques to bargain on the house and make offers on it. More specifically, a special crossover operator is used to determine a new offer, which is between a previously accepted offer and a previously rejected one. This is reasonable because an acceptable offer means it might too high and thus it could be reduced further; while a rejected offer means it might to be too low and thus should be adjusted higher. In this way, the system can find accommodation as cheap as possible for its student users. Moreover, since the evolutionary search is in parallel with the process of bargaining, sellers (landlords) can leave the accommodation renting market and the system would not re-start the negotiation process. So, our system is robust in such a dynamic situation. In addition, a special mutation operator (i.e., an offer of current generation minus a random number) is used to catch suddenly low prices that have suddenly appeared in the market. This could happen when a new landlord comes into the market with an unusually low price or exiting landlords suddenly reduce the price for quick sale. Currently, few web services for accommodation renting can monitor the market changing and carry out bargaining for a ‘good deal ’ on behalf of students. | |||||||||||||
Publication details | |||||||||||||
| |||||||||||||