Sharif Digital Repository

Clustering has great applications in various fields such as marketing, health, insurance, bioinformatics and etc. The assumption of regular parametric clusters is a common problem in current popular methods. This assumption is not valid in most real applications and greatly increases the clustering errors even to an unacceptable rate. Inspired by sardine fish, in this thesis we propose a new model with high elasticity factor that can cluster data without cluster shape constraints. This method uses the sardine fish model to augment clustering space dimension in order to achieve greater separability. The proposed method had some problems that were fixed and the final algorithm was finalized.... 

“Motor Theory” of speech perception discusses the role of motor competence in speech perception. Motor  Competence  is  accessed,  simulating  the  gestures  which  lead  to  the  heard  voice  signals.  Simulating is done using mirror neurons which share similar feelings between sender and receiver. Empathy,  caused  by  usage  of  mirror  neurons,  probably  caused  some  sort  of  link  between  inward  motor gestures and outward semantic categories at the origin of language. Some degree of linkage between sound and meaning is what we call “Sound Symbolism”. Along  with  describing  the  link  between  Motor  Theory  and  Sound  Symbolism,  this  thesis  took  two  approaches to analyze... 

This paper develops a novel insight and procedure that includes a variety of algorithms for finding the best solution in a structured multi-agent system with internal communications and a global purpose. In other words, it finds the optimal communication structure among agents and the optimal policy in this structure. First, a unique reinforcement learning algorithm is proposed to find the optimal policy of each agent in a fixed structure with non-linear function approximation like artificial neural networks (ANN) and eligibility traces. Secondly, a mechanism is presented to perform self-organization based on the information of the learned policy. Finally, an algorithm that can discover an... 

Identifying anomalous events is one of the vital topics in research as it often leads to the detection of actionable and critical information such as intrusions, faults, and system failures. With its importance, there has been a substantial body of work for network anomaly detection using supervised and unsupervised machine learning techniques with their own strengths and weaknesses. In this work, we take advantage of both worlds of unsupervised and supervised learning methods. The basic process model we present in this paper includes (i) clustering the training data set to create referential labels, (ii) building a supervised learning model with the automatically produced labels, and (iii)... 

The estimation of soil parameters in geotechnical practice is always an important and challenging task for the geotechnical engineer. Obtaining undisturbed samples from sands is generally very difficult and expensive, and in some cases impractical. A good prediction of sands parameters from insitu tests such as Cone Penetration Test (CPT) is one of the most challenging problems in geotechnical engineering. Using Calibration Chambers, a soil with predefined parameters is tested by cone penetrometer and some relationships are developed between CPT results and soil parameters. Using these relationships, in-situ results are interpreted.
In this thesis, after introducing previously developed... 

Control Charts are the most popular monitoring tools for profiles. The time that a control chart gives an out-of-control signal is not the real time of change. The actual time of change is called the change point. This study suggests two new algorithms to find the change (shift) in parameters of nonlinear profiles. First, using ordinary Kriging method, new points are estimated. Then, with the help of Bernoulli hypothesis test, the probability of detecting the change for new points is tested. Nonlinear profiles in this study follow the exponential family of distributions; in particular, Exponential, Poison and Gaussian distribution structures are used as nonlinear profiles. The proposed... 

Privacy preserving data mining helps organizations and companies not only to deal with privacy concerns of customers and regular limitations, but also to benefit from collaborative data mining. Utilizing cryptographic techniques and secure multiparty computation (SMC) are among widely employed approaches for preserving privacy in distributed data mining. The general purpose of secure multiparty computation protocols to compute specific functions on private inputs of parties in a collaborative manner and without revealing their private inputs. Providing rigorous security proof of secure multiparty computation makes it a good choice for privacy preservation, despite of its cryptographic... 

Cellular learning automata (CLA) is a distributed computational model that is introduced recently. This model is combination of cellular automata (CA) and learning automata (LA) and is used in many applications such as image processing, channel assignment in cellular networks, VLSI placement, rumor diffusion and modeling of commerce networks, and obtained acceptable results in these applications. This model consists of computational units called cells and each cell has one or more learning automata. In each stage, each automaton chooses an action from its actions set and applies it to the environment. Each cell has some neighboring cells that constitute its local environment. The local rule... 

The gamma cameras, based on scintillation crystal followed by an array of photomultiplier tubes (PMTs), play a crucial role in nuclear medicine. The use of square PMTs provides the minimum dead zones in the camera. The camera with squared PMTs also reduces the number of PMTs relative to the detection area.In this thesis, we introduced a new read-out method whereby the total cost of the read-out board will be decreased by a factor of 2.4; in return, the energy and spatial resolution of the system will be reduced by 0.3% and 0.4% respectively. We also implemented a positioning module in the FPGA chip via that the transmission rate between FPGA and the computer will be tripled, and the...