Sharif Digital Repository

The ability to accurately determine temporal relations between events is an important task for several natural language processing applications such as Question Answering, Summarization, and Information Extraction. Since current supervised methods require large corpora, which for many languages do not exist, we have focused our attention on approaches with less supervision as much as possible. This paper presents a fully generative model for temporal relation extraction based on the expectation maximization (EM) algorithm. Our experiments show that the performance of the proposed algorithm, regarding its little supervision, is considerable in temporal relation learning  

Submodularity is a discrete domain functional property that can be interpreted as mimicking the role of well-known convexity/concavity properties in the continuous domain. Submodular functions exhibit strong structure that lead to efficient optimization algorithms with provable near-optimality guarantees. These characteristics, namely, efficiency and provable performance bounds, are of particular interest for signal processing (SP) and machine learning (ML) practitioners, as a variety of discrete optimization problems are encountered in a wide range of applications. Conventionally, two general approaches exist to solve discrete problems: 1) relaxation into the continuous domain to obtain an... 

Automatic extraction of temporal relations between event pairs is an important task for several natural language processing applications such as Question Answering, Information Extraction, and Summarization. Since most existing methods are supervised and require large corpora, which for many languages do not exist, we have concentrated our efforts to reduce the need for annotated data as much as possible. This paper presents two different algorithms towards this goal. The first algorithm is a weakly supervised machine learning approach for classification of temporal relations between events. In the first stage, the algorithm learns a general classifier from an annotated corpus. Then,... 

Parallel corpora are the necessary resources in many multilingual natural language processing applications, including machine translation and cross-lingual information retrieval. Manual preparation of a large scale parallel corpus is a very time consuming and costly procedure. In this paper, the work towards building a sentence-level aligned English-Persian corpus in a semi-automated manner is presented. The design of the corpus, collection, and alignment process of the sentences is described. Two statistical similarity measures were used to find the similarities of sentence pairs. To verify the alignment process automatically, Google Translator was used. The corpus is based on news... 

In almost all of the currently working circuits, especially in analog circuits implementing signal processing applications, basic arithmetic operations such as multiplication, addition, subtraction and division are performed on values which are represented by voltages or currents. However, in this paper, we propose a new and simple method for performing analog arithmetic operations which in this scheme, signals are represented and stored through a memristance of the newly found circuit element, i.e. memristor, instead of voltage or current. Some of these operators such as divider and multiplier are much simpler and faster than their equivalent voltage-based circuits and they require less... 

The ability to accurately classify temporal relation between events is an important task for a large number of natural language processing applications such as Question Answering (QA), Summarization, and Information Extraction. This paper presents a weakly-supervised machine learning approach for classification of temporal relation between events. In the first stage, the algorithm learns a general classifier from an annotated corpus. Then, it applies the hypothesis of "one type of temporal relation per discourse" and expands the scope of "discourse" from a single document to a cluster of topically-related documents. By combining the global information of such a cluster with local decisions... 

The Conventional Fire Flooding (CFF) process application feasibility on fractured carbonated reservoirs remained questionable. In this paper first combustion parameters and reaction kinetics of a naturally fractured low permeability carbonated heavy oil reservoir in Iran called Kuh-E-Mond applied to simulation study. After that, simulator has been validated with Kuh-E-Mond combustion tube experiment. Recovery mechanism in single block matrix is different from one in conventional model since oxygen first flows into the fractures and then diffuses from all sides into the matrix. Combustion of the oil in the fractures produces some water ahead of fracture combustion front which prohibits oxygen... 

Predicting the degree of compositionality of noun compounds such as snowball and butterfly is a crucial ingredient for lexicography and Natural Language Processing applications, to know whether the compound should be treated as a whole, or through its constituents, and what it means. Computational approaches for an automatic prediction typically represent and compare compounds and their constituents within a vector space and use distributional similarity as a proxy to predict the semantic relatedness between the compounds and their constituents as the compound's degree of compositionality. This paper provides a systematic evaluation of vector-space reduction variants across kinds, exploring... 

Object tracking is a difficult and primary task in many video processing applications. Because of the diversity of various video processing tasks, there exists no optimum method that can perform properly for all applications. Histogram-based particle filtering is one of the most successfu1 object tracking methods. However, for dealing with visual tracking in real world conditions (such as changes in illumination and pose) is still a challenging task. In this paper, we have proposed a color-based adaptive histogram particle filtering method that can update the target model. We have used the Bhattacharyya coefficients to measure the likelihood between two color histograms. Our experimental... 

The goal of image segmentation is to find regions that represent objects or meaningful parts of objects. In this paper a new method is presented for color image segmentation which involves the ideas used for community detection in social networks. In the proposed method an initial segmentation is applied to partition input image into small homogeneous regions. Then a weighted network is constructed from the regions, and a community detection algorithm is applied to it. The detected communities represent segments of the image. A remarkable feature of the method is the ability to segments the image automatically by optimizing the modularity value in the constructed network. The performance of... 

In this work, an optimum frequency is found for the operation of single cavity photonic switches. At this optimum point, the transmission contrast of ON and OFF states takes its highest value, while keeping the device power threshold relatively low and the device speed acceptably high. Then, the dynamic behavior of a typical single cavity all optical switch is investigated in the optimum operation point through temporal Coupled Mode Theory. Switching speed and power are discussed, and the device is shown to be applicable for telecommunication and data processing applications. The analysis is quite general, and can be used for resonant structures, such as photonic crystals and microring... 

Image registration is a basic task in medical image processing applications like group analysis and atlas construction. Similarity measure is a critical ingredient of image registration. Intensity distortion of medical images is not considered in most previous similarity measures. Therefore, in the presence of bias field distortions, they do not generate an acceptable registration. In this paper, we propose a sparse based similarity measure for mono-modal images that considers non-stationary intensity and spatially-varying distortions. The main idea behind this measure is that the aligned image is constructed by an analysis dictionary trained using the image patches. For this purpose, we use... 

Sigma-Delta modulator ADCs used in signal processing applications are usually implemented by switched-capacitor (SC) circuits and CMOS transmission gates. Clock feed-through effect is one of the main non-ideal parameters existing in SC integrators degrading modulator total SNDR and its linearity. In this paper, a comprehensive analysis of clock feed-through effect on CMOS transmission gates on both rising and falling edges on output node will be presented. The main interferer parameters such as clock signal timing model, input signal level and switch parameters effect on output error will be analyzed. Finally, circuit simulations using 0.18um CMOS technology in ADS environment show the... 

Skin detection is a difficult and primary task in many image processing applications. Because of the diversity of various image processing tasks, there exists no optimum method that can perform properly for all applications. In this paper, we have proposed a novel skin detection algorithm that combines color and texture information of skin with cellular learning automata to detect skin-like regions in color images. Skin color regions are first detected, by using a committee structure, from among several explicit boundary skin models. Detected skin-color regions are then fed to a texture analyzer which extracts texture features via their color statistical properties and maps them to a skin...