Authors’ names extraction from scanned documents

Due to the excellent performance of the HMM (hidden Markov model) in pattern recognition, it has been widely used in voice recognition, text recognition. The HMM has also been applied to the intrusion detection. The intrusion detection method based on the HMM is more efficient than other methods. The HMM based intrusion detection method is composed by two processes: one is the HMM process; the other is the hard decision process, which is based on the profile database. Because of the dynamical behavior of system calls, the hard decision process based on the profile database cannot be efficient to detect novel intrusions. On the other hand, the profile database would consume many computer resources. For these reasons, the combined detection method was provided in This work. The neural network is a kind of artificial intelligence tool and is combined with the HMM to make soft decision. In the implementation, radial basis function model is used, because of its simplicity and its flexibility to adapt pattern changes. With the soft decision based on the neural network, the robustness and accurate rate of detection model are greatly improved. The efficiency of this method has been evaluated by the data set originated from New Mexico University.

When a digital library user searches for publications by an author name, she often sees a mixture of publications by different authors who have the same name. With the growth of digital libraries and involvement of more authors, this author ambiguity problem is becoming critical. Author disambiguation (AD) often tries to solve this problem by leveraging metadata such as coauthors, research topics, publication venues and citation information, since more personal information such as the contact details is often restricted or missing. In this paper, we study the problem of how to efficiently disambiguate author names given an incessant stream of published papers. To this end, we propose a "BatchAD+IncAD" framework for dynamic author disambiguation. First, we perform batch author disambiguation (BatchAD) to disambiguate all author names at a given time by grouping all records (each record refers to a paper with one of its author names) into disjoint clusters. This establishes a one-to-one mapping between the clusters and real-world authors. Then, for newly added papers, we periodically perform incremental author disambiguation (IncAD), which determines whether each new record can be assigned to an existing cluster, or to a new cluster not yet included in the previous data. Based on the new data, IncAD also tries to correct previous AD results. Our main contributions are: (1) We demonstrate with real data that a small number of new papers often have overlapping author names with a large portion of existing papers, so it is challenging for IncAD to effectively leverage previous AD results. (2) We propose a novel IncAD model which aggregates metadata from a cluster of records to estimate the author's profile such as her coauthor distributions and keyword distributions, in order to predict how likely it is that a new record is "produced" by the author. (3) Using two labeled datasets and one large-scale raw dataset, we show that the proposed method is much more efficient than state-of-the-art methods while ensuring high accuracy.

This paper describes the information properties of museum specimen labels and machine learning tools to automatically extract Darwin Core (DwC) and other metadata from these labels processed through Optical Character Recognition (OCR). The DwC is a metadata profile describing the core set of access points for search and retrieval of natural history collections and observation databases. Using the HERBIS Learning System (HLS) we extract 74 independent elements from these labels. The automated text extraction tools are provided as a web service so that users can reference digital images of specimens and receive back an extended Darwin Core XML representation of the content of the label. This automated extraction task is made more difficult by the high variability of museum label formats, OCR errors and the open class nature of some elements. In this paper we introduce our overall system architecture, and variability robust solutions including, the application of Hidden Markov and Naïve Bayes machine learning models, data cleaning, use of field element identifiers, and specialist learning models. The techniques developed here could be adapted to any metadata extraction situation with noisy text and weakly ordered elements.

This paper proposes an approach combining the set of Krawtchouk moments and HMMs (Hidden Markov Models) for unconstrained offline handwritten Chinese character recognition. The Krawtchouk moments can be effectively used as pattern features in analysis of two-dimensional images. Since the basis set is orthogonal in the discrete domain of the image coordinate space, the implementation of Krawtchouk moments doesn't involve any numerical approximation or coordinate transformation. This property makes Krawtchouk moments superior to conventional, more continuous moments such as Legendre moments and Zernike moments. In the field of character recognition, HMMs architecture has been used widely because of its strong adaptability to written distortion and variability. In this paper, Krawtchouk moments and HMMs are combined to recognize unconstrained offline handwritten Chinese amount in words. Experiments show that the average recognition of Krawtchouk is higher than Zernike and Legendre moments within the HMM framework.

In this paper we present an end-to-end trainable optical character recognition (OCR) system for recognizing machine-printed text in Thai documents. The end-to-end OCR system is based on a script-independent methodology using hidden Markov models. Our system provides an integrated workflow beginning with annotation and transcription of training images to performing OCR on new images with models trained on transcribed training images. The efficacy of our end-to-end OCR system is demonstrated by rapidly configuring our OCR engine for the Thai script. We present experimental results on Thai documents to highlight the specific challenges posed by the Thai script and analyze the recognition performance as a function of amount of training data.

There are more than 6,000 bibliographic styles around the world. The most common academic writing and citation styles have emerged since the beginning of the 20th century. Chicago style (The Chicago Manual of Style, CMOS, CMS) was introduced by the University of Chicago in 1906, and its 17th edition was released in 2017. The APA style (The Publication Manual of the American Psychological Association, APA) was published in 1952, and the 7th edition was released in 2019. Most university students in Turkey cannot be fully taught how to use one of the citation styles, such as Chicago and APA. When master's and doctoral theses completed in Turkey are examined in terms of compliance with citation formats, it is seen that most theses are not written in these formats. However, it is considered necessary to be used in thesis writing guides. The same is true for academic publications in the types of articles and books written by faculty members. Few academic books in Turkey are published in accordance with these styles. While some Institutes or publishers require academic works to be written in Chicago, APA, or other citation styles, it is clear that these styles are not generally taught to students. Several reasons can be listed for this result. Styles such as Chicago and APA and academic publications, which were written in English, have been prepared as guidelines. Turkish students or academics can access the full texts of these styles in English either by purchasing the book format or by subscribing to the electronic versions of these styles by their universities. These two access options are not common in our country, as they require payment of a fee. If there are Turkish translations of styles such as Chicago and APA, using them is also an option. For example, it is possible to obtain the Turkish translation of the 6th edition, the old version of the APA, in PDF format free of charge. However, even translations of older editions of other citation styles are sold as commercial products and are not freely available. The information in styles such as Chicago and APA is updated, and new editions are made available over time. The current version of the APA, the 7th edition, was published in English with 428 pages in 2019, and the 17th edition of Chicago was published in 2017 with 1144 pages in English. As of August 2022, they do not have Turkish translations. Even if they are translated into Turkish, students can only access their full texts by purchasing them. Researchers who submit their work to an institute or journal that requests the use of Chicago and APA styles usually prepare their studies by looking at the examples of one of the different editions they find on the internet. In this case, researchers are often unaware of which editions of Chicago style, which has 17 editions, or APA, which has seven editions, they are looking at. If universities require styles such as Chicago or APA for dissertation writing, Turkish universities should subscribe to them and make them freely available to their students, as do European and US universities. However, access to their English texts will not be a complete solution. These styles, prepared to write academic texts in English, will need to be translated into Turkish and made available to students. Whenever these styles are available to students and researchers in full text, awareness will increase that Chicago and APA styles do not only contain 3-5 pages of information on footnote and bibliography. These styles contain all the basic information students and researchers should know about academic research and writing. This is why APA's English has 428 pages and Chicago style has 1144 pages. The styles, such as Chicago and APA, cannot be learned correctly by students in Turkey due to the difficulty of accessing the full English texts of these styles and the Turkish translations of the current editions. Thus, supervisors, editors, and publishers, who have to check and correct Turkish academic texts, face many negative consequences. Due to the lack of knowledge of academic writing and citation, academic ethical violations can be made by students and researchers unintentionally. To eliminate all these negativities, the ISNAD Citation System was developed with the project support of Sivas Cumhuriyet University in 2018. ISNAD is a Turkey-based academic writing and citation system developed in Turkish for use in academic studies. It will also be effective in preventing ethical violations if students learn how to conduct scientific research and cite sources correctly from a system written in Turkish in their mother tongue. To achieve this result, it is necessary to provide students with an accessible style where they can learn academic research and writing. The ISNAD Citation System has been prepared and used for this purpose. Access to the full text and web version of ISNAD is completely free. English, Arabic, and Persian translations of ISNAD are also available free of charge. It will also be translated into Azerbaijani, Kyrgyz, Uzbek, and other Turkish languages. ISNAD can also be used with Citation management software such as Zotero, EndNote, and Citavi. Seminars are held free of charge, and lecture videos are broadcast on ISNAD Academy TV to popularize the use of ISNAD and learn how to use it with academic writing tools. User support is also provided to students 24/7 via the ISNAD website. Universities, publishers, and researchers can use ISNAD free of charge, which is offered with an open-access license. As of August 2022, 37 universities and 136 peer-reviewed journals are using ISNAD. ISNAD is more accessible to students and easier to learn. Citation styles differ according to the discipline. Disciplines that use up-to-date quantitative data such as questionnaires and interviews prefer to cite (in-text system/Author-Date) by specifying the publication date immediately after the author's surname. The branches of science trying to reach the oldest written sources, such as manuscripts, archival documents, and classical Works, cite the source name after the author's name (system with footnotes/Author-Title). ISNAD has both in-text and footnote versions. However, in the academic texts written in both versions, the works are written in the same order and format in the bibliography. Therefore, ISNAD is suitable for all branches of science. Preparing the bibliography in a single format allows the bibliography data to be presented as “clean data.” Accepting ISNAD as a reference system by universities and journals will also contribute to the standardization of academic writing. Thus, "clean metadata" will be provided to academic databases and indexes using bibliographic data. Currently, the metadata of academic publications based in Turkey is not very qualified and not visible enough from abroad. In reaching the goal of adopting Turkish as the language of science, it is already necessary to have a citation style developed in Turkish. ISNAD will continue to update itself with the opinions and suggestions of students and academicians and will remain open public at no charge.

Devanagari is a popular old script used by multiple major languages of the Indian sub-continent such as Hindi, Marathi, Nepali and Sanskrit. The popularity of the script is self-evident from the fact that more than 500 million use it. Complexity associated with unconstrained Devanagari writing is more than English cursive due to the possible variations in the order number, direction and shape of constituent strokes. In this paper, we propose a scheme for standardization of stroke order thereby training the user and using the strategy for future isolated Devanagari character recognition in iPhone. On the basis of manual study of stroke classes, one Hidden Markov Models (HMM) is created for each class which is chosen for character recognition in that HMM class. Standardization of stroke order reduces the number of HMM promoting higher probability and faster recognition of isolated Devanagari character. A few characters among the 20 most occurring characters are taken as samples and their manual study generates certain stroke classes. The second phase deals with the training and recognition of isolated Devanagari character on the basis of the HMM classes. Though there has been some work on Optical Character Recognition (OCR) of Devanagari character, recognition on a smartphone platform like iPhone is a new and promising field.

Historical newspapers are invaluable resources that capture the cultural, social, economic, and political dimensions of their time, providing unique insights into past events and daily life. However, the potential of these archives remains underutilised due to the limitations of traditional research methods, which rely on laborious manual examination of individual pages or articles. Digital archives have begun to address this issue by enhancing accessibility, but the lack of effective tools for parsing and analysing the contents still restricts their utility for large-scale historical research. Optical Character Recognition (OCR) technology plays a foundational role in converting scanned images into searchable text. However, OCR alone cannot adequately handle the complexities of historical newspapers, which often feature degraded image quality, inconsistent typography, and intricate layouts. In parallel, the effective segmentation and ordering of newspaper contents is essential for generating well-structured and logically ordered data from specific newspaper sections. This paper focuses on the development of a pipeline that addresses these challenges, using the daily newspaper Le Sémaphore de Marseille as a case study (35,703 issues covering the period 1827 to 1944). The proposed system integrates layout analysis, OCR, and Generative AI-driven information extraction techniques to extract specific data elements from scanned images, namely ship arrivals data, and to store them in a machine-readable format for further analysis. This well-structured output enables historians to analyse historical trade patterns, economic trends, and regional interactions more efficiently and at a much larger scale. An evaluation of the key pipeline components demonstrates the effectiveness of the system, achieving an F1-score of 96% in paragraph segmentation and information extraction, while also highlighting the need for further improvement in the critical OCR component.

Off-line recognition of handwritten Korean and alphanumeric characters using hidden Markov models

The technology of Optical Character Recognition (OCR) is used to generate texts in the process of digitizing print documents. Usually these texts need to be indexed and organized to simplify their access and retrieval. One of the powerful approaches in accomplishing this task is the use of Automated Text Classification. However, it is currently impossible for OCR technology to recognize all characters with an accuracy of 100%. We therefore propose the use of combined linguistic features in automated classification of OCR texts to formulate an informative feature set. The proposed method was experimentally evaluated using Japanese OCR texts. Empirical results indicate that the combination of linguistic features improved classification performance of OCR texts.

This paper presents fast and accurate method for recovering diacritical markings and guessing original meaning of the word from the context based on a hidden Markov model and the Viterbi algorithm. The proposed algorithm might find usage in any area where erroneous text might appear, such as a web search engine, e-mail messages, office suite, optical character recognition or helping to type on small mobile device keyboards.

Text detection and recognition in scene images and videos attract much attention in computer vision recently. However, most existing text detection and recognition methods only focus on static images. In this paper an end-to-end scene text recognition method based on multi frame tracking is proposed for text in videos, in which temporal information is employed to improve performance. First, an end-to-end text recognition method based on a unified deep neural network is used to detect and recognize text in each frame of the input video. Then, multi frame text tracking is employed through associations of texts in current frame and several previous frames to obtain final results. Experiments on ICDAR datasets demonstrate that the proposed method outperforms the state-of-the-art methods in end-to-end video text recognition.

Optical Characters Recognition (OCR) has been an active subject of research since the early days of computers. Despite the age of the subject, it remains one of the most challenging and exciting areas of research in computer science. In recent years it has grown into a mature discipline, producing a huge body of work. Arabic character recognition has been one of the last major languages to receive attention. This is due, in part, to the cursive nature of the task since even printed Arabic characters are in cursive form. This paper describes the performance of combining Hough transform and Hidden Markov Models in a multifont Arabic OCR system. Experimental tests have been carried out on a set of 85.000 samples of characters corresponding to 5 different fonts from the most commonly used in Arabic writing. Some promising experimental results are reported.

Tools, techniques and protocols necessary for building up digital libraries have now fully evolved with computing power that allows parallel processing, multitasking, parallel consultation and parallel knowledge navigation and sohare tools that facilitate artificial intelligence and interadivity. Coincided with availability of sofhvare, hardware and networking technology, the advent of world wide web 0its, ever incr easing usage and highly evolved browsers have paved the way for creation of digital libraries. With rapid developments in technologies necessary for developing digital libraries, the world of digital information resources has expanded rapidly and exponentially. Increasing number of publishers are using the internet as a global way to offer their publications to the international community of scientists and technologists. With the technology available at an effordable cost, the libraries are initiating small digitisation projects as an individual library or as a group of libraries. Building-up digital collection and the infrastructure required to access them is a challenge that every library has to deal with. The article delves into technological evolution, cultural revolution and contents enrichment that led to revolution in growth and development of digital libraries. The artide has two distinct parts, while the first part deals with various aspects of building, accessing and organising digital resources and collection, the second part elaborates on the process, technology, formats, compression techniques and tools used in digital imaging. The article describes optical character recognition (OCR) and advocates for hybrid solution for preservation of digital information. http://dx.doi.org/10.14429/dbit.21.6.3547

In this paper, we propose a novel approach for online handwriting recognition (HR) based on hidden Markov model (HMM). In a conventional HMM-based HR system, the input test sample is recognized by first measuring the log-likelihood score from each class-specific HMM, and then the class with the highest score is assigned as the recognized class. It is observed that, for a given test sample, the difference in log-likelihood scores of top-2 outputs (classes) is often less for faithful classification. The problem intensifies for those scripts that have a large set of similar shape characters such as the Indic script. To address this problem, first, we analyze the HMM states corresponding to the top-2 classes and identify a subset of states that most discriminate the two classes. Afterwards, the final recognition among the two classes is carried out by comparing the log-likelihood scores of these chosen states. Since the proposed methodology focuses only on the most discriminative states of the two classes, therefore it enhances the classification confidence as well as overall recognition accuracy with least added complexity. The proposal is demonstrated for character and limited vocabulary word recognition tasks and evaluated on the locally collected Assamese character and word databases. The experimental results are promising over the conventional HMM-based HR system.