Sharif Digital Repository

In the last decade, Intrusion Detection Systems has attracted attention due to their importance in network security, but still they've shortcomings. Generating a lot of low level alerts is the main problem. Many of these alerts are actually false positives. One suggested solution is Alert Correlation Analysis. Because of false positives alert correlation techniques are not able to build accurate scenarios, but the accuracy of alerts can be verified with the aid of the information logged in the host systems. In this dissertation after surveying the current alert correlation techniques, a model will be introduced to effectively verify the generated alerts and to apply correlation techniques to... 

Semantic relation extraction is one of the challenging information extraction’s steps. The main task of relation extraction is finding of relationships in text corpus using Machine Learning methods. There are plenty of usage for relation extraction such as providing information for Question Answering systems, protein interaction detection in biomedical corpora and ontology population. There are many research take placed in this area in which relation extraction task is defined as a classification problem and in most of them, this problem is solved by SVM method. Results of recent research imply that current state of relation extraction methods are far from appropriate state that is... 

In this thesis the goal is to use data Mining techniques for finding features of different Partial discharges happen in power transformer insulation defect models in order to monitor the insulation condition of in-service power transformers continuously and on-line and to identify any type of insulation defect in power transformers at the early stage of formation.For this purpose, power transformer Insulation defects are modeled physically and partial discharge current pulses are recorded as Partial discharge data.Tow format of data are investigated, in first one, recorded data in one power frequency cycle are explored and in second one, individual PD pulses are considered for study.In... 

Networked computers are expanding more and more around the world, and digital social networks becoming of great importance for many people's work and leisure. Emails, Weblogs and Instant Messengers are popular instances of social networks. In this thesis, the main target is to have an advertisement according to user favorites and interests by mining his/her interactions in digital social networks. Briefly, in our method social network users are categorized based on the topics exchanges between them in the network, these topics discovered by mining of flowing data in that environment, considering that these topics shows the user willing, finally relevant advertisements will be represented to... 

Abnormality detection and prognosis of epileptic patients with artificial intelligence and machine learning techniques is still in its early experimental stages. Surgical candidacy determination for epilepsy depends on the clinical actions which involve an intracranial electrode implantation followed by prolonged electrographic monitoring (EEG phase II) .This invasive test is very costly, painful and time consuming. Here the goal is integration of the two following paradigms: 1-Non invasive multimodality data of epilepsy. 2- Artificial intelligence and machine learning techniques. We have used human brain multi-modality database system that includes patient’s demographics, clinical and EEG... 

Nowadays, the world is based on the interaction between individuals, groups and different systems. The actual networks that have a complex structure and behavior are called complex networks. Complex networks are one of the new knowledge that studies the connections. The complex systems represented as graph, with non-trivial topological features—features that do not occur in simple networks.With the vast development of computer networks, complex networks appear in different categories such as social networks, citation networks, collaboration networks and communication networks. Data mining is the process of exploring hidden knowledge in data bases and it has applications in complex networks.... 

Today, machine learning has many applications in various industries, and healthcare is not an exception. Machine learning algorithms are used for medical diagnosis, make predictions about patients’ future health, newly-discovered treatment effect on patients prediction, drug recommendation system, build risk models and survival estimators and health risk prediction models. One of the topics that has received less attention in the world, especially in Iran, is the prediction of the surgery duration. This is very important because operating rooms in hospitals are the primary source of hospital revenue; We also need to predict the duration of surgery as accurately as possible in order to... 

Recently, data scientists have been paying much attention to sports. Many researches have been done in this field, using data mining and machine learning techniques. The following research aims to predict the results of football matches, which consists of two general approaches. For the first and second approaches, we used video game data and match statistics, respectively. In both approaches, it was tried to predict not only the final result (win, draw, or loss) but also the final goal difference. In the first approach, the home team victory was predicted by 73% accuracy, the draw by 75.4%, and the home team defeat by 73.7%. Nevertheless, in the second approach, the home team victory was... 

Due to the application of data streams in various data sources such as Web click streams, Web pages, and data generated by sensors and satellites, data streams have attracted a huge attention recently. A data stream is an ordered sequence of points that must be accessed in order and can be read only once or a small number of times. For mining such data, the ability to process in one pass along with limited memory usage is very important. Data stream clustering also has received a huge attention in recent years and numerous algorithms are developed in this field. None of them has paid attention to the feature selection problem as an effective factor in clustering quality especially when the...