Word sense disambiguation is the task of identifying the correct sense for the Word in a given context among a finite set of possible senses, and plays an important role in many natural language processing applications such as machine translation, document classification, and information retrieval.In this paper we proposed to use LDA Topic Model for disambiguating Farsi homograph words with new features. A Topic Model is a statistical Model for extract Topics from documents of a corpus. We use Latent Dirichlet Allocation (LDA) that is an unsupervised technique.The system achieved 97% precision for 4 high frequently Farsi homograph words.

Information Technology (IT), management and industrial engineering are correlated academic disciplines whose publications have risen signi ficantly over the last decades. The aim of this study is to analyze the research evolution, determine the important Topics and areas, and depict the trend of interdisciplinary Topics in these domains. To accomplish this, text mining techniques are used and a combination of bibliographic analysis and a Topic Modeling approach are applied to relevant publications in the Web of Science (WoS) repository over the last 20 years. In the Topic extraction process, a heuristic function was suggested for key extraction, and some new applicable criteria were defi ned to compare the Topics. Moreover, a novel approach was proposed to determine the high-level category for each Topic. The results determined the hot-important Topics, and incremented, decremented and  fixed Topics are identi fied. Subsequently, a comparison between high-level research areas con firmed strong scienti fic relationships between them. This study presents a deep knowledge about the internal research evolution of domains and illustrates the effect of Topics on each other over the past 20 years. Furthermore, the methodology of this study could be applied to determine interdisciplinary Topics and observe the research evolution of other academic domains.

A probabilistic Topic Model assumes that documents are generated through a process involving Topics and then tries to reverse this process, given the documents and extract Topics. A Topic is usually assumed to be a distribution over words. LDA is one of the first and most popular Topic Models introduced so far. In the document generation process assumed by LDA, each document is a distribution over Topics and each word in the document is sampled from a chosen Topic of that distribution. It assumes that a document is a bag of words and ignores the order of the words. Probabilistic Topic Models such as LDA which extract the Topics based on documents-level word co-occurrences are not equipped to benefit from local word relationships. This problem is addressed by combining Topics and n-grams, in Models like Bigram Topic Model (BTM). BTM modifies the document generation process slightly by assuming that there are several different distributions of words for each Topic, each of which correspond to a vocabulary word. Each word in a document is sampled from one of the distributions of its selected Topic. The distribution is determined by its previous word. So BTM relies on exact word orders to extract local word relationships and thus is challenged by sparseness. Another way to solve the problem is to break each document into smaller parts for example paragraphs and use LDA on these parts to extract more local word relationships in these small parts. Again, we will be faced with sparseness and it is well-known that LDA does not work well on small documents. In this paper, a new probabilistic Topic Model is introduced which assumes a document is a set of overlapping windows but does not break the document into those parts and assumes the whole document as a single distribution over Topics. Each window corresponds to a fixed number of words in the document. In the assumed generation process, we walk through windows and decide on the Topic of their corresponding words. Topics are extracted based on words co-occurrences in the overlapping windows and the overlapping windows affect the process of document generation because; the Topic of a word is considered in all the other windows overlapping on the word. On the other words, the proposed Model encodes local word relationships without relying on exact word order or breaking the document into smaller parts. The Model, however, takes the word order into account implicitly by assuming the windows are overlapped. The Topics are still considered as distributions over words. The proposed Model is evaluated based on its ability to extract coherent Topics and its clustering performance on the 20 newsgroups dataset. The results show that the proposed Model extracts more coherent Topics and outperforms LDA and BTM in the application of document clustering.

Research in the field of video surveillance systems has been improving because of the increasing need for intelligent monitoring, control and management. Given the large amount of data on these intelligent transportation systems, extracting patterns and automatically labeling them is a challenging task. In this paper, a Topic Model was used to detect and extract traffic patterns at intersections so that visual patterns are transformed into visual words. The input video is first split into clips. Then, the flow characteristics of the clips, which are based on abundant local motion vector information, are computed using optical flow algorithms and converted to visual words. After that, with a non-probabilistic Topic Model, the traffic patterns are extracted to the designed system by a group sparse Topical coding method. These patterns represent visible motion that can be used to describe a scene by answering a behavioral question such as: Where does a vehicle go? The results of the implementation of the proposed method on the QMUL video database show that the proposed method can correctly detect and display meaningful traffic patterns such as turn left, turn right and crossing a roundabout.

Probabilistic Topic Models have been very popular in automatic text analysis since their introduction. These Models work based on word co-occurrence, but are not very flexible with respect to the context in which cooccurrence is considered. Many probabilistic Topic Models do not allow for taking local or spatial data into account. In this paper, we introduce a probabilistic Topic Model that benefits from an arbitrary-length co-occurrence window and encodes local word dependencies for extracting Topics. We assume a multinomial distribution with Dirichlet prior over the window positions to let the words in every position have a chance to influence Topic assignments. In the proposed Model, Topics being shown by word pairs have a more meaningful presentation. The Model is applied on a dataset of 2000 documents. The proposed Model produces interesting meaningful Topics and reduces the problem of sparseness.

With the proliferation of the Internet and the rapid growth of electronic articles, text classification has become one of the key and important tools for data organization and management. In text classification a set of basic knowledge is provided to the system by learning. Then, new input documents enter to one of the subject groups. In health literature due to wide variety of Topics, preparing such a set of early education is a very time consuming and costly task. The purpose of this article is to present a hybrid Model of learning (supervised and unsupervised) for the subject classification of health scientific products that performs the classification operation without the need for an initial labeled set. To extract the thematic Model of health science texts from 2009 to 2019 at PubMed database, data mining and text mining were performed using machine learning. Based on Latent Dirichlet Allocation Model, the data were analyzed and then the Support Vector Machine was used to classify the texts. In the findings of this study, the Model was introduced in three main steps. In data preprocessing, the unnecessary words were eliminated from the data set and the accuracy of the proposed Model increased. In the second step, the themes in the texts were extracted using the Latent Dirichlet Allocation method, and as a basic training set in step 3, the data were backed up by the Support Vector Machine algorithm and the classifier learning was performed with the help of these Topics. Finally, with the help of the classification, the subject of each document was identified. The results showed that the proposed Model can build a better classification by combining unsupervised clustering properties and prior knowledge of the samples. Clustering on labeled samples with a specific similarity criterion merges related texts with prior knowledge, and the learning algorithm teaches classification by supervisory method. Combining classification and clustering can increase the accuracy of classification of health texts.

Automatic Topic detection seems unavoidable in social media analysis due to big text data which their users generate. Clustering-based methods are one of the most important and up-to-date categories in Topic detection. The goal of this research is to have a wide study on this category. Therefore, this paper aims to study the main components of clustering-based-Topic-detection, which are embedding methods, distance metrics, and clustering algorithms. Transfer learning and consequently pretrained language Models and word embeddings have been considered in recent years. Regarding the importance of embedding methods, the efficiency of five new embedding methods, from earlier to recent ones, are compared in this paper. To conduct our study, two commonly used distance metrics, in addition to five important clustering algorithms in the field of Topic detection, are implemented by the authors. As COVID-19 has turned into a hot trending Topic on social networks in recent years, a dataset including one-month tweets collected with COVID-19-related hashtags is used for this study. More than 7500 experiments are performed to determine tunable parameters. Then all combinations of embedding methods, distance metrics and clustering algorithms (50 combinations) are evaluated using Silhouette metric. Results show that T5 strongly outperforms other embedding methods, cosine distance is weakly better than other distance metrics, and DBSCAN is superior to other clustering algorithms.

Metalepsis, in its narratological sense, is a trope in which an unnatural relationship is built between different levels of narrative. The natural relationship between narrative levels is formed by the act of narrating; a character from one level becomes the narrator of another. The term “ Metalepsis of Topic and illustration” can be coined to name a similar trope. This trope has been used for centuries in Persian poetry. Every Image comprises a Topic and an illustration. The Topic is what is being talked about and the illustration is what the Topic is compared to. When several images are present along together, two different levels are distinguishable: the level of the Topic and the level of illustration. The natural relationship between these two levels is similarity and any other relationship will result in metalepsis. As in the narratological metalepsis, there is always a paradox in the metalepsis of Topic and illustration. The effect of these two kinds of metalepsis is also similar and can be humorous, fantastic, or a mixture of the two.