درخت حوزه‌های تخصصی

Learning effective segment depictions and resemblance measures are fundamental to the recuperation execution of a substance based picture recuperation (CBIR) structure. Regardless of wide research tries for a significant long time, it stays one of the most testing open gives that broadly impedes the achievements of real-world CBIR structures. The key test has been credited to the extraordinary "semantic hole" subject that happens between low-level photo pixels got by technologies and raised close semantic thoughts saw by a human. Among various techniques, AI has been successfully analyzed as a possible course to interface the semantic gap in the whole deal. Impelled by late triumphs of significant learning techniques for PC vision and various applications, in this paper, we try to address an open issue: if significant learning is a longing for spreading over the semantic gap in CBIR and how much updates in CBIR endeavors can be cultivated by exploring the front line significant learning methodology for learning feature depictions and likeness measures. Specifically, we explore a structure of significant learning with application to CBIR assignments with a wide game plan of definite examinations by investigating front line significant learning methodologies for CBIR endeavors under moved settings. From our exploratory examinations, we find some encouraging results and compress some huge bits of information for upcoming research.

We have a huge civic project, includes several sub-projects, which are divided into activities. This project is a project to rehabilitate 550,000 hectares of agricultural land in the provinces of Khuzestan and Ilam. The project has been divided according to the plots of land. The Jehad in Tehran is managing the projects. They need project scheduling as well as resource levelling and "lot-sizing". Levelling and lot sizing are the most important issues in utilizing the limited resources. For determining the scope and the size of those sub-projects as well as their parallel activities, so far, many models have been proposed. However, the models are weak either in higher resource utilization or in solving numerically the problems. In this paper, our effort has been concentrated on developing scheduling, resource levelling, and lot sizing model, based on balancing utilization of resources, so that the real size civil project could be solved within an acceptable duration time. This paper proposes a goal programming linear model for simultaneous project scheduling and resource levelling. This model determines the best schedule of sub-activities (optimal "lot size" of each sub-activity) to reach the minimum amount of diversion of resources consumed from the number of resources available for the entire periods of the planning horizon. In fact, if the best "lot sizes" have been taken, then minimum fluctuation of the active resources is reached. The proposed model has been used to schedule a project with 87 activities. This project has been scheduled and, accordingly, the optimal volume (the "lot size") of sub-activities have been determined for each activity at any period of time. In this way, only 4 resources out of a total of 32 resources are in shortage. In contrast, the scheduling of this project, using the CPM, results in a shortage of 13 resources.

Being aware of the waiting time for selling residential units is one of the important issues in the housing sector for the majority of people, especially investors. There are several factors affecting the waiting time for selling residential units. Determining the influential factors on the time period of selling real estates can lead to an informed decision making by real estate consultants, sellers as well as those seeking to buy real estates. Using a real estate database in Iran, the present paper proposes a two-module procedure. The first module deals with implementation of association rule mining. Using the well-known association rule mining techniques namely FP-Growth, several association rules have been extracted which indicate the effective factors on the waiting time for selling residential units. Generated association rules have been evaluated based on metrics such as support, confidence and lift and finally the best rules are selected. The main objective of the second module is to develop a fuzzy inference system which can determine the factors influencing the waiting time for selling residential units from historical data, so that the model can be used to estimate the time it to sell the property for a real estate agency. Several IF-THEN rules are extracted from this module. Extracted rules can be used by real estate agencies as well as buyers and sellers of residential units to make better decisions in their investments. In conclusion section, a number of suggestions for future studies are presented. For example, machine learning algorithms such as neural networks, decision trees, etc. can also be used to predict the duration of residential units’ sale. The main objective of the second module is to develop a fuzzy inference system which can learn about the factors that influence the waiting time for selling residential units from historical data, so that the model can be used to estimate the time it takes to sell the property for a real estate agency. Several IF-THEN rules are extracted from this module. Extracted rules can be used by real estate agencies as well as buyers and sellers of residential units to make better decisions in their investments.