In this paper, we show that the problem of grammar induction could be modeled as a combination of several model selection problems. We use the infinite generalization of a Bayesian model of cognition to solve each model selection problem in our grammar induction model. This Bayesian model is capable of solving model selection problems, consistent with human cognition. We also show that using the notion of history-based grammars will increase the number and decrease the complexity of model selection problems in our grammar induction model. This results in the induction of a better grammar which leads to 9.1 points increase in F1 measure, for parsing the section 22 of Penn treebank in comparison with a similar model that does not use history-based grammar induction techniques.

Classical image deconvolution seeks an estimate of the true image when the blur kernel or the point spread function (PSF) of the blurring system is known a priori. However, blind image deconvolution addresses the much more complicated, but realistic problem where the PSF is unknown. Bayesian inference approach with appropriate priors on the image and the blur has been used successfully to solve this blind problem, in particular with a Gaussian prior and a joint maximum a posteriori (JMAP) estimation. However, this technique is unstable and suffers from significant ringing artifacts in various applications. To overcome these limitations, we propose a regularized version using $H^1$ regularization terms on both the sharp image and the blur kernel. We present also useful techniques for estimating the smoothing parameters.  We were able to derive an efficient algorithm that produces high quality deblurred results compared to some well-known methods in the literature.

This paper provides an in-depth evaluation of various supervised machine learning models used for predicting diabetes. It discusses the strengths and limitations of several algorithms, including Decision Trees, Random Forest, Rotation Forest, Ensemble Classifier, K-Star, Simple Bayes, Logistic Regression, Functional Tree, and Perceptron Neural Network. The study utilizes a publicly available diabetes dataset from chistio. ir, which includes 520 samples, comprising 200 diabetic patients and 320 non-diabetic patients, and assesses 16 features. Results are validated on the Weka 3. 6 open-source platform, using metrics such as AUC, classification accuracy (CA), F1 score, precision, and recall.

Over recent decades, there has been a growing interest in semi-supervised clustering. Compared to the supervised or unsupervised clustering methods for solving different real-life problems, reviewed articles show that semi-supervised clustering methods are more powerful, and even a small amount of supervised information can significantly improve the results of unsupervised methods. One popular method of incorporating partial supervised information is through labeled data. In this study, we propose a semi-supervised clustering algorithm called ConvexClust. The proposed method improves data clustering using a geometric view borrowed from the Lune concept in the connectivity index and 10% of labeled data. Clustering starts with the use of labeled data and the formation of a convex hull. It continues over the labeling of non-labeled data and the updating of the convex hull in an iterative process. Evaluations of three UCI datasets and sixteen artificial datasets show that the proposed method outperforms the other semi-supervised and traditional clustering techniques.

The volume of Farsi information on the Internet has been increasing in recent years. However, most of this information is in the form of unstructured or semi-structured free text. For quick and accurate access to the vast knowledge contained in these texts, the information extraction methods are essential to generate knowledge bases. In recent years, relation extraction as a sub-task of information extraction has received much attention. While many of these systems were developed in English and other well-known languages, the systems for information extraction in Farsi have received less attention from researchers. In this systematic research for semi-automatic relation extraction, Persian Wikipedia articles were presented as reliable and semi-structured sources. In this system, the relation extraction is performed with the assistance of patterns that are automatically obtained with an approach based on distant supervised. In order to apply the distant supervised, the vast knowledge base of Wikidata has been used as a source in perfect synchronization with Wikipedia. The results show that the average precision value for all relations is 76. 81%, which indicates an enhancement of precision compared to other methods in Farsi.

In some instances, the occurrence of an event can be influenced by its spatial location, giving rise to spatial survival data. The accurate and precise estimation of parameters in a spatial survival model poses a challenge due to the complexity of the likelihood function, highlighting the significance of employing a Bayesian approach in survival analysis. In a Bayesian spatial survival model, the spatial correlation between event times is elucidated using a geostatistical model. This article presents a simulation study to estimate the parameters of classical and spatial survival models, evaluating the performance of each model in fitting simulated survival data. Ultimately, it is demonstrated that the spatial survival model exhibits superior efficacy in analyzing blood cancer data compared to conventional models.

In this paper we briefly review some of the Bayesian techniques for sample size determination in different trials. The two main areas are inferential and decision theoretic frameworks. In the inferential approach we are usually concerned with inference about unknown parameter(s) of interest and sample sizes are determined by taking the parameters of posterior distribution into account. In the decision theoretic approach the problem is treated as a decision problem and using a proper utility function the optimal sample size is determined by optimizing an objective function

Keywords are the main focal points of interest within a text, which intends to represent the principal concepts outlined in the document. Determining the keywords using traditional methods is a time consuming process and requires specialized knowledge of the subject. For the purposes of indexing the vast expanse of electronic documents, it is important to automate the keyword extraction task. Since keywords structure is coherent, we focus on the relation between words. Most of previous methods in Persian are based on statistical relation between words and didn’ t consider the sense relations. However, by existing ambiguity in the meaning, using these statistic methods couldn’ t help in determining relations between words. Our method for extracting keywords is a supervised method which by using lexical chain of words, new features are extracted for each word. Using these features beside of statistic features could be more effective in a supervised system. We have tried to map the relations amongst word senses by using lexical chains. Therefore, in the proposed model, “ FarsNet” plays a key role in constructing the lexical chains. Lexical chain is created by using Galley and McKeown's algorithm that of course, some changes have been made to the algorithm. We used java version of hazm library to determine candidate words in the text. These words were identified by using POS tagging and Noun phrase chunking. Ten features are considered for each candidate word. Four features related to frequency and position of word in the text and the rest related to lexical chain of the word. After extracting the keywords by the classifier, post-processing performs for determining Two-word key phrases that were not obtained in the previous step. The dataset used in this research was chosen from among Persian scientific papers. We only used the title and abstract of these papers. The results depicted that using semantic relations, besides statistical features, would improve the overall performance of keyword extraction for papers. Also, the Naive Bayes classifier gives the best result among the investigated classifiers, of course, eliminating some of the features of the lexical chain improved its performance.

Graph based semi-supervised methods for automatic image annotation are mainly focused on single-label problems. However, most of the real world problems require multiple labels per image. As a hybrid semi-supervised approach, LGC+ML-KNN is proposed for multi-label image annotation. LGC is a graph based semi-supervised learning algorithm that annotates unlabeled samples. Subsequently, ML-KNN learns from many more labeled samples, as compared to the initial training set. Experiments on several datasets confirm that the proposed approach has better accuracy than available methods, especially when a very small portion of the training set are the labeled samples.