Sharif Digital Repository

Keyphrase extraction, as an important open problem of Natural Language Processing (NLP), is useful as a stand-alone task in the field of Information Extraction and as an upstream task for Information Retrieval, text summarization and classification,etc. In this study, regarding the needs in Persian NLP, artificial neural networks are adopted to extract keyphrases from single documents and a graph-based re-scoring method is proposed for multidocument keyphrase extraction. The proposed method for extracting keyphrases from multiple documents consists of two steps: (1) extracting keyphrases of each document in a cluster using a sequence to sequence model with attention, and (2) re-scoring the... 

This project has been conducted in two related phases: In the first phase, we have attempted to write a program capable of answering thematic similarity multiple-choice questions without utilizing any training data. The best performance in this phase was attained by the 25-topic LDA model using the Hellinger distance between the probability distributions of the poetic verses. This model managed to attain an accuracy of 42%, which is very close to the average human performance of 43%. In the second phase, two tasks of seven-class classification and binary classification were defined based on the p-value of the questions. To this end, the questions were initially ranked according to the... 

In recent years, word embeddings as the word representation have captured the attention of natural language processing (NLP) researches. One of the great advantages of word embeddings is their capability in representing the relationships of the words. Therefore, using word embeddings in NLP applications results in better performance.Despite widespread attention towards word embedding in late years, Persian word embeddings have not achieved sensible progress. One of the Persian word embeddings difficulties is related to that, Persian is a low-resource language in comparison with worldwide languages. Therefore, Persian word embedding quality is lower than English. Consequently, the accuracy of... 

News plays a very important role in stock market trading. Nowadays news from a different part of the world and about different fields can be accessed easily, and for a successful trade, it is necessary to analyze accurately and use this big data and information as soon as possible. For this reason, this thesis tries to present and study models based on Deep Learning networks and Natural Language Processing for financial news analysis and predicting stock indices movement. This research takes advantage of a language model for learning and representing news text, and beside this language model it uses deep learning networks at multiple levels to extract proper features from each news in a day... 

In contrast to automatic speech recognition (ASR), which converts a speaker’s spoken utterance into a text string, spoken language understanding (SLU) is aimed at interpreting user’s intentions from their speech utterances. Traditionally, this has been accomplished by writing context-free grammars (CFGs) or unification grammars (UGs) manually. The manual grammar authoring process is laborious and expensive, requiring much expertise. In addition, robustness is a vital requirement of these modules, because the input of these modules comes from a speech recognition unit and always contains errors. In recent years, many data-driven models have been proposed for spoken language understanding, but... 

Improvement of automatic speech recognition systems and the growth of audio data (such as broadcast news, voice mail, telephony conversations and meetings) have attracted plenty of research interest in the field of speech summarization. The goal of this dissertation is to improve the performance of the speech summarization in the domain of multi-party conversations, specifically meetings. Most of the previous work in this field are inheritted from the text summarization counterpart, whithout paying much attention to the discourse specific information of the multi-party conversations. The main idea of this work is to use discourse information to improve the accuracy of extracted summaries in... 

The main objective of all managers of the economic organizations, manufacturing and industrial companies and service providers is to take steps toward improvement and effective use of different resources like workforce, capital, materials, energy and information. To attain the desired productivity, managers have to observe requirements such as establishment of a working organizational structure, effective implementation methods, safe equipment and tools, balanced working atmosphere and qualified human force as the most important factor. Therefore productivity culture should be publicized throughout the organization, and human force should be considered as the main component of this process.... 

Considering the advantages of optical communications and code division multiple access technique, Optical Code Division Multiple Access Networks (OCDMANs) was introduced. Thence, many researchers have suggested methods for creating Quality of Service (QoS) differentiated classes in OCDMANs. In this thesis, we have comprehensive review on the QoS assignment methods and enhance them by offering new techniques. In physical layer, we concentrate on variable chip rate OCDMANs and in MAC sub-layer, a new algorithm named Adaptive Level Control (ALC) is introduced. Then, we focus on the physical design of OCDMANs. Firstly, we solve the problem of one-class OCDMAN and introduce few simple and... 

In this study, the cooperative inertial navigation systems are used to prevent the error growth of the inertial navigation system in unmanned aerial vehicles that fly in group. Here, the navigation problem is defined for a group of unmanned vehicles, flying with full six degrees of freedom and the objective is to estimate the navigation information of all cooperators. Due to the nonlinear equations, the decentralized extended Kalman filter is used to estimate the states of the system. This means that there is a Kalman filter in each vehicle. The inputs of this filter are the range and the bearing angles between the vehicle and the cooperators, the estimated position of the cooperators and... 

In this study, integrated navigation systems that utilize velocity-matching algorithm and are used in automatic landing of Unmanned Aerial Vehicles (UAV) are investigated. At first, the accuracy of the navigation system in the landing phase is analyzed and the relationship between the accuracy of inertial sensors, flight parameters and the accuracy of the navigation system is derived analytically. Then, by integrating the dynamic model of the UAV and the inertial measurement unit, a new algorithm for navigation of fixed-wing UAVs is presented. The integration mechanism of the proposed algorithm is tightly coupled; hence, it is capable to eliminate the errors of UAV model and inertial system.... 

Consistent growth in volume and dynamic behavior of traffic mandates new requirements for fast and adaptive resource allocation in modern optical networks.Elastic optical networks have been proposed to provide a competitive and scalable solution by creating tunable high-capacity optical communication networks in which network elements are configured according to the actual need to minimize resource usage for the specified service requirements. Resource allocation with quality of service constraints is one of the most challenging problems in elastic optical networks which is conventionally formulated as a complex mixed-integer nonlinear optimization program.In this research work, we focus on... 

The perpuse of this thesis is to design and fabricate a laboratory one degree of freedom stable platform under turbulences by model-based design method. To do this, different parts of the system were first modeled in the Solidworks software and then fabricated with various tools. With proper assembly of electronic and mechanical components, the original body of the system was created. Then, the controller motor with a stable platform which is attached to it was modeled and simulated in MATLAB software. Then, the parameters of the equations were calculated using standard identification tests and TCACS optimization algorithm. One of the challenges of this thesis is the malfunction of the... 

In this research, a cooperative terminal guidance algorithm has been derived for simultaneous multi-interceptor collision with a target. This algorithm is obtained using the combination of model predictive control method and differential game theory. One of the features of this method is the calculation of the optimal guidance command in the form of an explicit relationship. Another feature of this method is its resistance to disturbances due to the use of differential game theory. The first innovation of this research is the use of this hybrid method to create a guidance algorithm, even in the non-cooperative guidance problem. The next innovation is in creating a cooperative terminal... 

Mobile edge computing (MEC) is a promising technology that aims to resolve cloud computing’s issues by deploying computation resources at the edge of mobile network and in the proximity of users. The advantages of MEC include reduced latency, energy consumption, and load on access and mobile core networks, to name but a few. Despite all the aforementioned advantages, the mobility of mobile network users causes the traditional MEC architecture to suffer from several issues, such as decreased efficiency and frequent service interruption. One of the methods to manage users’ mobility is virtual machine (VM) migration, where the VM containing the user’s task is migrated to somewhere closer to... 

Hybrid engines are one of a variety of chemical propulsion systems that have been given special attention in recent years by various industries such as space and defense. The most important reasons for considering this type of system are its high safety, its affordable price compared with liquid fuel propulsion systems, the possibility of switching on and off and changing thrust. This system has a lot of disadvantages as well. This system cannot meet all the needs of different industries due to the low regression rate and, consequently, the limitation of the operating range, thrusts and special impulse. As a result, increasing the fuel regression rate in these engines can greatly expand... 

Optical impairments, as a challenging obstacle in optical communication systems, have been under deep attention of optical engineers in recent years. These impairments can be divided into two categories of linear and nonlinear impairments. Dispersion is a significant linear impairment leads to broadening of the optical signal and interference between transmitted symbols during detection. Nonlinear impairments such as cross-phase modulation (XPM) and self-phase modulation (SPM) also distort the transmitted signal through the fiber, causing interference at the receiver and degrading the system's performance. An appropriate solution for compensating optical fiber impairments is to utilize... 

The performance evaluation involves a process by which the employees are periodically and officially evaluated. Major reasons of performance evaluation include, among others, identification of efficient employees and granting bonuses to them to create stimuli to improve their and other employees' performances. The main objective of performance evaluation constitutes study of people's performances in the organization in order for directors to take proper and required decisions to enhance both quantities and qualities of the employees' works. The performance evaluation is deemed to be a major critical function of the management of human resources to facilitate organizational effectiveness. In... 

One of the proposed methods for enhancing optical networks and increasing their capacity is the use of the band multiplexing concept. This approach leverages additional available bands in optical fibers, enabling a significant boost in network capacity at much lower costs compared to other methods. However, fully utilizing the potential of this concept requires efficient resource allocation strategies to adequately counteract the nonlinear effects in optical fibers caused by increased bandwidth. In this research, we introduce an innovative method that simultaneously considers the physical layer constraints and user quality of service requirements for resource allocation in multi-band optical... 

The performance of visible light communication (VLC), as a key technology for next-generation wireless communication networks, 6G, can be significantly enhanced through the integration of metasurface-based optical reconfigurable intelligent surfaces (RIS). This research work introduces a novel channel model for a VLC system incorporating RIS-equipped transmitters, where the transparent optical RIS is directly integrated with the array of LEDs in VLC transmitter, enabling beam-steering. To harness the beam-steering capabilities of the modeled RIS-equipped transmitter, a traffic-aware resource allocation scheme is proposed and efficiently solved using a tailored fractional programming...