Markup Language Research Articles

Development of a XLM-Encoded Machine-Readable Dictionary for Yoruba Word Sense Disambiguation

The development of the disambiguation component of a Yorùbá to English machine translation system is hindered by several factors. One of these is the lack of machine readable sense inventory for ambiguous Yorùbá words. This study addressed the problem by developing a machine readable dictionary for ambiguous Yoruba verbs. To achieve this, ambiguous Yorba verbs and their translations were collected from existing bilingual dictionaries. The collected lexicons were transformed into machine readable format using the Extensible Markup Language (XML) Format. The accuracy of translation of the machine readable dictionary was evaluated using mean opinion score, with a score of 4.37 over the scale of 5. This study covered the total number of ninety-three (93) monosyllabic verbs with two hundred and forty-one (241) senses, which gives a coverage of 69.5% of the ambiguous monosyllabic verbs in Yoruba Language. The sense inventory was also used as a component of a Yoruba Word Sense Disambiguation system, and an accuracy of 94.6% was achieved. This study concludes that the digitized data can increase the accuracy of Word Sense Disambiguation component of a Yorùbá to English machine translation system.

A Two-Stage Long Text Summarization Method Based on Discourse Structure

This paper proposes a two-stage automatic text summarization method based on discourse structure, aiming to improve the accuracy and coherence of the summary. In the extractive stage, a text encoder divides the long text into elementary discourse units (EDUs). Then a parse tree based on rhetorical structure theory is constructed for the whole discourse while annotating nuclearity information. The nuclearity terminal nodes are selected based on the summary length requirement, and the key EDU sequence is output. The authors use a pointer generator network and a coverage mechanism in the generation stage. The nuclearity information of EDUs is to update the word attention distribution in the pointer generator, which not only accurately reproduces the critical details of the text but also avoids self-repetition. Experiments on the standard text summarization dataset (CNN/DailyMail) show that the ROUGE score of the proposed two-stage model is better than that of the current best baseline model, and the summary achieves corresponding improvements in accuracy and coherence.

A novel hybrid approach for text encoding: Cognitive Attention To Syntax model to detect online misinformation

Most approaches for text encoding rely on the attention mechanism, at the core of the transformers architecture and large language models. The understanding of this mechanism is still limited and present inconvenients such as lack of interpretability, large requirements of data and low generalization. Based on current understanding of the attention mechanism, we propose CATS (Cognitive Attention To Syntax), a neurosymbolic attention encoding approach based on the syntactic understanding of texts. This approach has on-par to better performance compared to classical attention and displays expected advantages of neurosymbolic AI such as better functioning with little data and better explainability. This layer has been tested on the task of misinformation detection but is general and could be used in any task involving natural language processing.

Expert System for Medical Predictions of Hepatitis B

Hepatitis B virus has been a major concern to the world as it has been a killer infection for many years. It is an infection that has adverse effects on the human liver causing cirrhosis, and hepatocellular carcinoma. In many countries, many persons had been infected in the past with this deadly disease, and it resulted in many deaths. This has been so, due to the high cost and time required to undergo laboratory checks and consult a medical doctor. Not so many people could afford the medical bills required for early detection of the virus. The researchers seek to curb the challenge with the development of an Expert System for Medical Predictions of Hepatitis B. The software is meant to accurately predict if a patient has hepatitis B or not, the extent of the infection if any, and also suggest treatment. This is to enable for early detection and treatment of infected patients. This reduces the cost and time required to get treatment for hepatitis B. The methodology adopted for the analysis of this work is Object Oriented Analysis and Design (OOAD) and Unified Modeling Language (UML). The program is written using JavaScript alongside HTML5 (Hyper Text Markup Language), CSS (Cascading Style Sheets) and bootstrap framework. The database is implemented using MySQL and the program is implemented using PHP (Hypertext Preprocessor). From the results, the Expert system has accurately diagnosed if a patient has chronic or acute hepatitis B and also prescribed suitable treatment.

BioNetGMMFit: estimating parameters of a BioNetGen model from time-stamped snapshots of single cells

Mechanistic models are commonly employed to describe signaling and gene regulatory kinetics in single cells and cell populations. Recent advances in single-cell technologies have produced multidimensional datasets where snapshots of copy numbers (or abundances) of a large number of proteins and mRNA are measured across time in single cells. The availability of such datasets presents an attractive scenario where mechanistic models are validated against experiments, and estimated model parameters enable quantitative predictions of signaling or gene regulatory kinetics. To empower the systems biology community to easily estimate parameters accurately from multidimensional single-cell data, we have merged a widely used rule-based modeling software package BioNetGen, which provides a user-friendly way to code for mechanistic models describing biochemical reactions, and the recently introduced CyGMM, that uses cell-to-cell differences to improve parameter estimation for such networks, into a single software package: BioNetGMMFit. BioNetGMMFit provides parameter estimates of the model, supplied by the user in the BioNetGen markup language (BNGL), which yield the best fit for the observed single-cell, time-stamped data of cellular components. Furthermore, for more precise estimates, our software generates confidence intervals around each model parameter. BioNetGMMFit is capable of fitting datasets of increasing cell population sizes for any mechanistic model specified in the BioNetGen markup language. By streamlining the process of developing mechanistic models for large single-cell datasets, BioNetGMMFit provides an easily-accessible modeling framework designed for scale and the broader biochemical signaling community.

Thermodynamic modeling with uncertainty quantification using the modified quasichemical model in quadruplet approximation: Implementation into PyCalphad and ESPEI

The modified quasichemical model in the quadruplet approximation (MQMQA) considers the first- and the second-nearest-neighbor coordination and interactions, particularly useful in describing short-range ordering (SRO) in complex liquids such as molten salts, slag in metal processing, and electrolytic solutions. The present work implements the MQMQA into the Python based open-source software PyCalphad for thermodynamic calculations. This endeavor facilitates the development of MQMQA-based thermodynamic database with uncertainty quantification (UQ) and propagation (UP) using the open-source software ESPEI. A new database structure based on Extensible Markup Language (XML) is proposed for ESPEI evaluation of MQMQA model parameters. Using the KF-NiF2, KCl-NaCl-MgCl2, and CaCl2-CaF2-LiCl-LiF salt systems as examples, we demonstrate the successful implementation of MQMQA in PyCalphad through thermodynamic calculations of Gibbs energy, equilibrium quadruplet fractions, and phase diagram, as well as database development with UQ and UP using ESPEI. Furthermore, as an application of the present implementation, both the LiF–TbF3 and LiF-HoF3 systems have been modeled by MQMQA for the first time, which are in good agreement with experiments. The present implementation hence offers an open-source capability for performing CALPHAD modeling for complex liquids with SRO using MQMQA plus a new XML database structure.

Implementation Chatbot for SMEs Using Artificial Intelligence Markup Language to Improve Customer Integration and Business Performance

Chatbots have been extensively adopted to produce more positive customer experiences as customers now spend more time in digital surroundings. Despite the technological advancement and benefits of chatbots for client service, exploration of chatbot operations for small and medium-sized enterprises (SMEs) is limited. The absence of exploration explaining the struggles faced by SMEs contributes to the gap in SMEs’ chatbot adoption. This exploration determines the features and rudiments that fit with SMEs’ characteristics and their guests' interactions with chatbots. A mixed-methods approach is used to understand SMEs’ needs. The first study uses interviews with SME business owners and their customers in order to explore the features that chatbots should offer for SME by identifying combinations between chatbots' generic features and SMEs' customer characteristics. The second study tests features identified in SMEs customers to empirically test featured chatbots’ influence on anthropomorphism, perceived enjoyment, perceived ease of use, perceived usefulness, and how they affect SMEs’ customer intentions to use chatbots and their shopping intentions. This paper contributes to the emerging service literature on the use of chatbots for service interactions, particularly for SMEs using Artificial Intelligence Markup Language. The data that has been obtained was analyzed using a Likert scale. As a result, the accuracy of using Artificial Intelligence Markup Language is 84.8%.

Progress with Repository-based Annotation Infrastructure for Biodiversity Applications

Progress with Repository-based Annotation Infrastructure for Biodiversity Applications

DirKorp

In this paper, we present recent developments on a new version (v3.0) of DirKorp (Korpus direktivnih govornih činova hrvatskoga jezika), the first Croatian corpus of directive speech acts developed for the purposes of pragmatic research. The corpus contains 800 elicited speech acts collected via an online questionnaire with role-playing tasks, a method of simulated communication that is implemented under pre-set conditions. This method is suitable for researching speech acts due to the ability to collect a great number of examples of such acts of equal propositional content and illocutionary purpose used in the same controlled situations. The presented situations are classified into two categories with regard to the relationship between the participants of the communication act: (1) situations involving interlocutors who are not in a familiar relationship; (2) situations involving interlocutors in a familiar relationship. Assignments of the two categories are organized into four pairs, asking respondents to share a speech act of similar propositional content. The respondents were 100 Croatian speakers, all undergraduate (63%) or graduate students (37%) of the Faculty of Humanities and Social Sciences (University of Zagreb). The corpus has been manually annotated on the speech act level, each speech act containing up to 14 features: (1) respondent ID, (2) familiarity/unfamiliarity, (3) utterance type, (4) directive performative verb in 1st person, (5) illocutionary force, (6) propositional content, (7) T/V form, (8) exhortative, (9) lexical marker of request, (10) lexical marker of apology, (11) lexical marker of gratitude, (12) honorific title, (13) grammatical mood, and (14) modal verb in 2nd person. It contains 12,676 tokens and 1,692 types. The corpus is encoded according to the TEI P5: Guidelines for Electronic Text Encoding and Interchange, developed and maintained by the Text Encoding Initiative Consortium (TEI). DirKorp is available for download under the CC BY-SA 4.0 license from GitHub in TEI format. We describe applied pragmatic annotation as well as the structure of the corpus.

Artificial Intelligence Stimulus Response System

The Stimulus Response System (SRS) is based on the concept of artificial intelligence (AI) based mark-up language. (AIML) and is used to train the system for recognizing the user input. By making use of the previously built knowledge base this system is trying to analyze the user input and it provides the response. The Chatter Bot’s, reported in the literature, and was used mainly the indexing technique. The method is strongly dependent on the contents that are stored in a file or a database. Thus there is a need to develop a new system that allows you to carry on conversations with the system as long as possible. Stimulus Response System creates and adds knowledge to a computer. As a result, the computer will appear as human. The main objective of the work is to design an interactive software package which make the computer simple by building content that induces the client to carry on interactive conversations with the system as long as possible.

Strong and Weak Supervision Combined with CLIP for Water Surface Garbage Detection

Water surface garbage has a significant impact on the protection of water environments and ecological balance, making water surface garbage object detection a critical task. Traditional supervised object detection methods require a large amount of annotated data. To address this issue, we propose a method that combines strong and weak supervision with CLIP (Contrastive Language–Image Pretraining) for water surface garbage object detection. First, we train on a dataset annotated with strong supervision, using traditional object detection algorithms to learn the location information of water surface garbage. Then, we input the water surface garbage images into CLIP’s visual encoder to obtain visual feature representations. Simultaneously, we train CLIP’s text encoder using textual description annotations to obtain textual feature representations of the images. By fusing the visual and textual features, we obtain comprehensive feature representations. In the weak supervision training phase, we input the comprehensive feature representations into the object detection model and employ a training strategy that combines strong and weak supervision to detect and localize water surface garbage. To further improve the model’s performance, we introduce attention mechanisms and data augmentation techniques to enhance the model’s focus and robustness towards water surface garbage. By conducting experiments on two water surface garbage datasets, we validate the effectiveness of the proposed method based on the combination of strong and weak supervision with CLIP for water surface garbage object detection tasks.

Dynamic Text: How to Effectively Embed Emotions in Text-based Computer-Mediated Communication

Despite being an effective mode of communication, Computer-Mediated Communication (CMC), particularly regarding text-based exchanges, still has inherent limitations compared to in-person communication. The objective of the current study is to review the previous studies which identified and addressed the challenges, including difficulty in embedding emotional information in text effectively. Previous studies demonstrated changes in text properties might deliver emotional information as well as the two phases of text encodings could be completed simultaneously by interaction designs. The study also suggests future studies to find the emotional meaning of text features and to develop novel applications using machine learning technologies.

Implementation of Chatbot System on Tourism Objects in Banyumas Regency with AIML and Chatterbot

Information technology can be applied to all fields, including tourism. Tourism object information media can be implemented into the chatbot system to make the information search process more flexible. Currently, searching for tourist spot information is often done manually; this makes tourist information services limited in time, while the need for tourism information must always be available. This research aims to build a chatbot system using Artificial Intelligence Markup Language (AIML) and ChatterBot methods. Both methods are accessed from libraries in Python using input in the form of natural language that has been processed into certain patterns. The pattern determination process is carried out by classifying a collection of questions on the chatbot using the Support Vector Machines (SVM) method. Then the classification is divided into five attributes, namely address, ticket price, facilities, description, and access. The SVM model built obtained an accuracy rate of 88%. Based on the testing results on both models that have been tested, the approach with AIML results in an accuracy rate in answering questions correctly of 90%, while ChatterBot is 40%.

Automation of Customer Support System (Chatbot) to Solve Web Based Financial and Payment Application Service

One of the most important features of any online service is the quality of its customer care. However, with the development of NLP tools, businesses are considering automated chatbot solutions to keep up with the increasing demand for their products and services. In view of this, the chatbot was developed using AIML java interpreter library Program AB which helps match input and output predefined in the AIML file. AIML (Artificial Intelligence Markup Language) was used to preprocess and train the bot using ready-made AIML file for FAQ questions. Also, vaadin was used to build a web UI to interact with the trained AIML bot. Finally, a google script was written to translate from any language to English for the bot to understand and send the response in the preferred language of the user. Findings showed that the response time of the bot is dependent of the network, as the design gave a score of 70%, 80%, 90% and 90% for load testing, stability, reliability testing and usability testing, respectively. Also, the bot is compatible with different operating systems, both for forward compatibility and backward compatibility having a score of 95%. The bot was able to answer customer questions, enquiries and complaints and the response time of the bot depends on the strength of the network since it is web based. Hence, the system provided a simple, cheaper, and durable customer financial and payment application service. Since chatbots cannot answer all questions, businesses that decide to implement them should ensure that they have enough protections in place against attacks and that routine requests are standardised to ensure optimal performance.

Strategies and Approaches for Generating Identical Extensive XML Tree Instances

In recent years, XML has become the de facto internet wire language. Data may be organized and given context with the use of XML. A well-organized document facilitates the transformation of raw data into actionable intelligence. In B2B1 applications, the XML data is sent and created. This implies the need for fast query processing on XML data. The processing of XML tree sample queries (XTPQ) that provide an efficient response (also known as sample matching) is a topic of active study in the XML database field.DOM (Parser) may be used to transform an XML document into a tree representation. Extensible Markup Language (XML) query languages like XPath and XQuery use tree samples (twigs) to express query results.XML query processing focuses mostly on effectively locating all instances of twig 1 samples inside an XML database. Numerous techniques for matching such tree samples have been presented in recent years. In this study, we survey recent developments in XTPQ processing. This summary will begin by introducing several algorithms for twig sample matching and then go on to provide some background on holistic techniques to process XTPQ.

Text-to-Image Generation using Generative AI

Abstract—This survey reviews text-to-image generation by using different approaches. One of the approaches identified in this study is Cross-modal Semantic Matching Generative Adversarial Networks (CSM-GAN), which is used to increase semantic consistency between text descriptions and synthesised pictures for fine-grained text- to-image creation. This includes other two modules, Text Encoder Module and Textual-Visual Semantic Matching Module. We further discussed about Imagen which is a text- to-image diffusion model with photorealism and deep language understanding, which is used on the COCO dataset. Lastly, we discussed about Text to image synthesis used to automates image generation using conditional generative models and GAN, enhancing artificial intelligence and deep learning. Based on these approaches we present a review of text to image generation using generative AI. Keywords— Generative AI, Diffusion model, Text-to- image, Imagen, CSM-GAN

A systematic review of progress on hepatocellular carcinoma research over the past 30 years: a machine-learning-based bibliometric analysis.

Research on hepatocellular carcinoma (HCC) has grown significantly, and researchers cannot access the vast amount of literature. This study aimed to explore the research progress in studying HCC over the past 30 years using a machine learning-based bibliometric analysis and to suggest future research directions. Comprehensive research was conducted between 1991 and 2020 in the public version of the PubMed database using the MeSH term "hepatocellular carcinoma." The complete records of the collected results were downloaded in Extensible Markup Language format, and the metadata of each publication, such as the publication year, the type of research, the corresponding author's country, the title, the abstract, and the MeSH terms, were analyzed. We adopted a latent Dirichlet allocation topic modeling method on the Python platform to analyze the research topics of the scientific publications. In the last 30 years, there has been significant and constant growth in the annual publications about HCC (annual percentage growth rate: 7.34%). Overall, 62,856 articles related to HCC from the past 30 years were searched and finally included in this study. Among the diagnosis-related terms, "Liver Cirrhosis" was the most studied. However, in the 2010s, "Biomarkers, Tumor" began to outpace "Liver Cirrhosis." Regarding the treatment-related MeSH terms, "Hepatectomy" was the most studied; however, recent studies related to "Antineoplastic Agents" showed a tendency to supersede hepatectomy. Regarding basic research, the study of "Cell Lines, Tumors,'' appeared after 2000 and has been the most studied among these terms. This was the first machine learning-based bibliometric study to analyze more than 60,000 publications about HCC over the past 30 years. Despite significant efforts in analyzing the literature on basic research, its connection with the clinical field is still lacking. Therefore, more efforts are needed to convert and apply basic research results to clinical treatment. Additionally, it was found that microRNAs have potential as diagnostic and therapeutic targets for HCC.

FindVehicle and VehicleFinder: a NER dataset for natural language-based vehicle retrieval and a keyword-based cross-modal vehicle retrieval system

Natural language (NL) based vehicle retrieval is a task aiming to retrieve a vehicle that is most consistent with a given NL query from among all candidate vehicles. Because NL query can be easily obtained, such a task has a promising prospect in building an interactive intelligent traffic system (ITS). Current solutions mainly focus on extracting both text and image features and mapping them to the same latent space to compare the similarity. However, existing methods usually use dependency analysis or semantic role-labelling techniques to find keywords related to vehicle attributes. These techniques may require a lot of pre-processing and post-processing work, and also suffer from extracting the wrong keyword when the NL query is complex. To tackle these problems and simplify, we borrow the idea from named entity recognition (NER) and construct FindVehicle, a NER dataset in the traffic domain. It has 42.3k labelled NL descriptions of vehicle tracks, containing information such as the location, orientation, type and colour of the vehicle. FindVehicle also adopts both overlapping entities and fine-grained entities to meet further requirements. To verify its effectiveness, we propose a baseline NL-based vehicle retrieval model called VehicleFinder. Our experiment shows that by using text encoders pre-trained by FindVehicle, VehicleFinder achieves 87.7% precision and 89.4% recall when retrieving a target vehicle by text command on our homemade dataset based on UA-DETRAC [1]. From loading the command into VehicleFinder to identifying whether the target vehicle is consistent with the command, the time cost is 279.35 ms on one ARM v8.2 CPU and 93.72 ms on one RTX A4000 GPU, which is much faster than the Transformer-based system. The dataset is open-source via the link https://github.com/GuanRunwei/FindVehicle, and the implementation can be found via the link https://github.com/GuanRunwei/VehicleFinder-CTIM.

Describing Inscriptions of Ancient Italy. The ItAnt Project and Its Information Encoding Process

This article discusses the challenges addressed in the digital scholarly encoding of the fragmentary texts of the languages of Ancient Italy according to the TEI/EpiDoc Guidelines in XML format. It describes the solutions and customisations that have been adopted for dealing with the peculiarities of our epigraphical documentation and with the formalisation of epigraphical information deemed interesting for data retrieval in a historical linguistic perspective. The making of a digital corpus consisting of new critical editions of selected inscriptions is a work carried out in the context of the project “Languages and Cultures of Ancient Italy. Historical Linguistics and Digital Models”, which aims to investigate the languages of Ancient Italy by combining the traditional methods, proper to historical linguistics, with methods and technology proper to the digital humanities and computational lexicography. More specifically, the purpose of the project is to create a set of interrelated digital language resources which comprise: (1) a digital corpus of texts editions; (2) a computational lexicon compliant with the Web Semantic requirements; (3) a relevant bibliographic reference dataset encoded according to the FRBRoo/LRMoo specifications. Additionally, selected textual data and scientific interpretations will be encoded using CIDOC CRM and its extensions, namely CRMtex and CRMinf. The present contribution thus tackles one of the main aspects of the project, and proposes significant innovations in the encoding of critical editions for epigraphic texts of fragmentary languages, which will hopefully foster future interoperability and integration with other external datasets, a paramount concern of the project.

Asiml-1553: A Markup Language for Mil-Std-1553B Avionics Bus Scheduler and Interface Control

Modern avionics is highly software intensive with various software components residing on different subsystems interfaced through avionics bus. Avionics system designer is responsible for defining the bus frame timings, message transmission schedules and details of interfaces. These are translated into design and finally into source code by the designers of individual subsystems. Growing complexity of avionics software coupled with frequent changes has necessitated use of highly layered architecture with tight partitions and open standards in order to obtain better maintainability, testability, portability and ease of upgrade. Textual configuration files offer benefit over embedding the configuration details into the code in terms of development, testing and certification efforts. XML is emerging as the worldwide standard for defining configuration files. This paper proposes ASIML-1553, a markup language for defining configuration files for avionics bus scheduler and interface control for MIL-STD1553B bus. Data from a typical combat aircraft are taken as case study to discuss the advantages ASIML-1553 offers in terms of turn-around time, quality and reliability. ASIML1553 can easily be adapted for interfaces other than MIL-STD-1553B such as ARINC-429 or Ethernet.