Applied and Computational Engineering

- The Open Access Proceedings Series for Conferences

Volume Info.

  • Title

    Proceedings of the 2023 International Conference on Machine Learning and Automation

    Conference Date

    2023-10-18

    Website

    https://2023.confmla.org/

    Notes

     

    ISBN

    978-1-83558-299-2 (Print)

    978-1-83558-300-5 (Online)

    Published Date

    2024-02-07

    Editors

    Mustafa İSTANBULLU, Cukurova University

Articles

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230462

    Construct a garbage recognition model using automatic machine learning based on EasyDL

    In modern times, machine learning has become an indispensable part of various industries. As the amount of data increases, reducing the time cost of manual annotation is crucial. AutoML emerges as a solution that effectively automates labor-intensive tasks like image annotation. In this article, we use Tencent's EasyDL to develop a garbage recognition function. The garbage recognition model completed through EasyDL achieved an average of over 90% in terms of accuracy and F1 score. This indicates that autoML can greatly reduce manual participation while ensuring a certain level of accuracy.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230463

    Syntax-aware bidirectional decoding Neural Machine Translation model

    The mainstream model in neural machine translation, the Transformer, relies heavily on self-attention mechanisms for translation operations. This approach has significantly improved both accuracy and speed. However, there are still some challenges. For instance, it lacks the incorporation of linguistic knowledge and the ability to leverage syntactic structure information in natural language for translation, leading to issues such as mistranslation and omission. Addressing the limitations of the Transformer's autoregressive decoding, which decodes from left to right without fully utilizing context information and is prone to exposure bias, this paper proposes a syntax-aware bidirectional decoding neural machine translation model. By employing both forward and backward decoders, the generated decoding results can encompass contextual information. Additionally, the model integrates dependency syntax to generate target language sentences with syntactic guidance. Finally, an optimization strategy involving the Teacher Forcing mechanism is introduced to balance the discrepancies between the Teacher Forcing training phase and the autoregressive testing phase, thus alleviating exposure bias issues.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230464

    Research advanced in Chinese word segmentation methods and challenges

    Chinese word segmentation refers to the process of dividing a sequence of Chinese characters into individual words. It constitutes a fundamental component of Chinese natural language processing. Due to the intricacies of the Chinese language, Chinese word segmentation has garnered significant attention from researchers. Based on a review of historical literature, segmentation methods can be broadly categorized into rule-based, statistical, semantic-based, and comprehension-based approaches. With the advancement of machine learning, neural networks have emerged as the mainstream algorithm for word segmentation. However, Chinese presents several unique challenges, leading to segmentation results that are less effective compared to morphological analysis in languages like English. Moreover, word segmentation faces new challenges such as dependency on the quality and scale of corpora, as well as domain-specific segmentation in diverse fields. Addressing these emerging challenges will undoubtedly become a focal point in future research endeavors in this field. This review provides a comprehensive summary of existing methods, discusses the current state of Chinese word segmentation, and outlines directions for addressing the evolving complexities in the field. As Chinese language processing continues to advance, finding robust solutions for accurate word segmentation remains a critical area of research.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230465

    Machine translation of classical Chinese based on unigram segmentation transformer framework

    In the translation work of Chinese ancient books, traditional manual translation is difficult and inefficient. As an important field of natural language processing, machine translation is expected to solve this problem. Due to the rapid development of NLP technology, prior works mainly follow the pipeline of Transformer when dealing with the machine translation task, which can extract the high-quality feature representation with its self-attention mechanism. The great success of Transformer has inspired the direction of our ancient text translation work. In this paper, we screen the Unigram word division by exploring and comparing, and propose a solution for the translation of ancient literary texts. Specifically, we adopt the evaluation of BLEU value and achieve the BLEU values of 43.4 and 40.03 for short and long sentences respectively. When compared with the results of Baidu Translation, our BLEU values increase by 8.12 and 5.18. Additionally, our translation results are more in line with the original text than Baidu Translation, demonstrating the potential and advantage of the model in bridging the ancient and modern Chinese era rift.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230466

    Decoding sentiment: A sentiment analysis model for movie reviews

    Sentiment analysis of movie reviews can provide valuable insights into movie reactions and preferences. To this end, this study proposes the Convolutional Long Short-Term Memory (ConvLSTM) neural network for movie review sentiment analysis. ConvLSTM can efficiently capture sequential information due to its recurrent neural network characteristics. Specifically, the movie review data are first tokenized. Next, the ConvLSTM analysis model is constructed additionally by fine-tuning its parameters to optimize the performance. The ConvLSTM model consists of multiple storage units that retain contextual information, enabling the model to identify long-distance dependencies in the text. The network is trained using a combination of positive and negative movie reviews, and the training process involves adjusting the model weights to minimize the classification error. Experimental results demonstrate the effectiveness of the proposed method in accurately predicting movie review sentiment. It outperforms traditional machine learning methods in sentiment analysis tasks. The findings demonstrate the potential of LSTM-based sentiment analysis in various applications such as movie recommendation systems and market research. This study's findings help advance the development of sentiment analysis techniques and are of great relevance in understanding and catering to audience preferences in the movie industry.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230467

    Analysis of Naive Bayesian and Back Propagation algorithms in iris classification

    An efficient taxonomy of irises can provide botanists with valuable tools. Machine learning algorithms can effectively improve the performance of iris classification models because they can automatically analyze and summarize data. To this end, this paper introduces Naive Bayesian (NB) and Back Propagation (BP) to build classification models. When creating the NB model, the petal and sepal data from the iris dataset are used sequentially as classification criteria to classify the data. When constructing the BP model, the author sets different iterations and outputs the loss function and accuracy of the BP model under different iterations. The study finds that the NB model has higher classification accuracy when using petal length and petal width as classification criteria, which is 17% higher than the classification accuracy using sepal length and sepal width. Therefore, the NB model is more suitable for classifying independent data. By studying the use of the BP algorithm to classify iris flowers, the automatic classification of iris flowers can be realized and the accuracy of classification can be improved. Compared with the traditional NB algorithm, the BP algorithm can better mine the hidden patterns and information in the iris data and make effective classifications. This study provides new insights and discoveries for the taxonomic study of Iris plants.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230468

    Prediction of diabetes progress based on machine learning approach

    Uropathy is a serious chronic disease whose prevalence is increasing at an alarming rate. Early detection and prediction of diabetes in women is important because of the increased risk of diabetes-related complications during pregnancy. This study introduces machine learning models to assess the likelihood of diabetes in women. The importance of studying characteristics and improving prediction accuracy to understand the nuances of categorization. Specifically, for data preprocessing, experiments are conducted to solve the problem of missing values and outliers by replacing the zero values of certain features with the median values of the corresponding features. This step reduces the impact of less reliable data on model performance. As recognition models, Gaussian Naive Bayes (GNB), Support Vector Machine (SVM), and Random Forest (RF) are built. Performance analysis is performed along with a careful exploration of the hyperparameter space. Scores for Receiver Operating Characteristic - Area Under the Curve (ROC-AUC) are used to compare various models. Different features affect the classification to different degrees. The experimental findings indicate that the modified random forest model demonstrates superior prediction accuracy and robustness. These findings can assist physicians in predicting a patient's risk of developing diabetes earlier.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230469

    Performance analysis of GFSK modulation over AWGN channel

    This paper analyses the performance of Gaussian Frequency Shift Keying (GFSK) modulation in Additive White Gaussian Noise (AWGN) channels. Using MATLAB 2022b for simulations, the study explores GFSK modulation principles, the role of Gaussian low-pass filters, the characteristics reflected by Power Spectral Density (PSD), and the influence of Bit Time-Bandwidth Product (BT) on Bit Error Rate (BER). It has been observed that the Gaussian filter restricts sidelobe amplitudes and enhances spectral efficiency. Through varying BT values, it is observed that higher BT values correlate with lower BER. Additionally, the study successfully reconstructs original baseband signals through sampling decisions, confirming GFSK modulation effectiveness.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230471

    Inclusive games: Accessible game design for the visually impaired

    Accessible game design is a key effort aimed at creating games that transcend physical and cognitive limitations. By carefully integrating a series of functions and strategies, designers empower players with various abilities to participate in the interactive field. Resolve visual barriers through high contrast visual effects, scalable fonts, and alternative color dependent information. Auditory elements are accompanied by visual cues or subtitles to ensure that the game narrative can be understood by auditory impairments. By utilizing customizable controls, motion sensitivity adjustments, and optional input methods, motion challenges are overcome. Recognize cognitive diversity by providing clear explanations, intuitive interfaces, and adjustable rhythms. This inclusive approach promotes innovation and encourages developers to explore new paths in game design, resonating with a wider range of users. As the gaming industry places increasing emphasis on usability, it emphasizes the industry's commitment to equal participation, while enhancing the creative process of each participant and enriching game gameplay.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230472

    Analysis of the prospective application of artificial intelligence in swimming

    Swimming has consistently maintained its status as a highly favored athletic pursuit for a span of one hundred years. In contemporary times, the burgeoning field of artificial intelligence (AI) has exhibited notable advancements, resulting in significant impacts across all domains. Examples of industries include finance, the service industry, and engineering. Furthermore, it has been implemented in various other sports previously. Presently, a predominant focus of scholarly inquiry is in the exploration of artificial intelligence (AI) applications within the realm of team sports, including disciplines such as basketball, volleyball, and rugby. Nevertheless, swimming shares certain characteristics with the aforementioned sports. In order to enhance the advancement of the swimming domain through the utilization of artificial intelligence (AI) technology, this essay will examine the feasibility of implementing certain AI applications that have been employed to support other sports, and will also introduce several AI technologies that have the potential to make distinctive contributions to the field of swimming.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230473

    Facial expression recognition with computer vision

    Facial Expression Recognition (FER) is a specialized field within the domains of computer vision and pattern recognition, which is dedicated to the automated identification and examination of facial expressions. Facial expression recognition (FER) has attracted considerable scholarly interest in recent years owing to its diverse array of applications and its potential ramifications across multiple disciplines, such as psychology, human-computer interaction, marketing, and security systems. The objective of this study is to present a thorough examination of the scholarly progression of FER, elucidating the significant achievements, approaches, and obstacles encountered by researchers in this domain. The study presents a selection of databases that are appropriate for Facial Expression Recognition (FER) and conducts a comparative analysis of these databases. The primary methodologies are examined, and recommendations are provided for each stage. In conclusion, this research presents several suggestions for addressing both obstacles and potential in future research endeavors.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230474

    Review of common spiking neural network simulation tools

    As a recognized third-generation neural network, the spiking neural network has high concurrency and complexity. Although the degree of research is still far from the previous generation of neural networks, spiking neural networks are excellent in performance and energy consumption. In this paper, the common spiking neural network simulation tools are reviewed. The most frequently used and mentioned tools are NEURON, NEST, and BRAIN. NEURON is more suitable for simulation based on biological applications, pays more attention to biological characteristics, and can support large-scale network simulation. Examples used in the official documentation are neural simulations of invertebrates and mammals. Large heterogeneous networks of point neurons or neurons with a few compartments are frequently simulated using NEST. In contrast to models that concentrate on the specific morphological and biophysical characteristics of individual neurons, NEST is appropriate for those that emphasize the dynamics, size, and structure of the nervous system. Brian was originally designed for research and teaching and is well suited as a teaching and presentation tool for simulating and observing the effects of different parameters for classical neural network projects such as picture classification. In addition, CSIM, SPLIT, SPINNAKER, and other tools also have their merits, but due to the low frequency of relevant references and lack of universality, this study will not give a detailed introduction.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230476

    Investigation of recent advances related to AI in mathematics education

    The integration of mathematics and Artificial Intelligence (AI) is an immensely promising field, making a comprehensive overview in this domain highly essential. This article provides an overview of two methodologies: Sofia and ITS. Sofia is an AI mathematical analysis model that engages in dialogues with multiple components and robots, performing calculations by processing sentences through Unix file system paths. ITS, on the other hand, is an intelligent tutoring system, comprising domain, student, tutor, and interface components, facilitating interactions within the teaching and learning processes. This article intricately delineates the implementation and functionalities of these two methodologies, encompassing their constituents, applications, and limitations. The applications of these methodologies are far-reaching in both mathematics and AI domains. For instance, in the realm of education, ITS can furnish personalized learning experiences by offering feedback and suggestions tailored to students' knowledge and behaviors. In the arena of mathematical analysis, Sofia can aid in analyzing and solving intricate mathematical problems, rendering support and guidance to students and professionals alike. In conclusion, this article offers a comprehensive overview of the amalgamation of mathematics and AI, introducing the implementations and applications of the Sofia and ITS methodologies. These approaches hold substantial potential within the fields of mathematics and AI, providing valuable references for future research and applications.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230477

    Investigation of recent advances in exercise-enhanced knowledge tracing models

    Exercising plays a significant role in Knowledge Tracing (KT), however, the majority of KT models struggle to effectively extract the abundant information embedded within students' exercise histories, leading to a prevalent issue of information erosion. Hence, it is vital for exploring methods to mine the data from the exercises to better forecast students' future performance. This paper is a review that aims to discuss recent improvements in KT models on mining information existing in exercises. This paper will briefly discuss DKT, a popular approach to knowledge tracing, and its limitation on mining information from exercises. And then, this review will introduce four exercise-enhanced knowledge tracing models including EERNN, EKT, HGKT and Concept-Aware Deep Knowledge Tracing. EERNN and EKT can track students’ knowledge acquisition by using vectors to represent knowledge concepts. HGKT goes a step further by examining the interconnectedness among different exercises based on EERNN and EKT. Concept-Aware DKT is an improvement of DKVMN, the other approach of KT, by considering the influence of knowledge concepts and corresponding exercises in more detail. Finally, the applications of exercise-enhanced KT models will be covered.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230479

    Research of types and current state of machine translation

    The background of machine translation dates back to the 1950s when scientists began exploring the use of computers for translation. Motivated by various challenges and needs, such as the high cost of manual translation and cross-cultural communication, machine translation has become a pivotal field. This overview delves into the research background, content, methods, key figures, conclusions, and future prospects of machine translation. It summarizes automatic evaluation metrics, corpus construction, and transfer learning, all of which contribute to enhancing translation performance. Currently, there are three mainstream categories of methods, which include rule-based translation, statistical translation, and neural network-based translation. The rule-based translation method relies on language rules and dictionaries for translation. Statistical machine translation involves the use of extensive bilingual corpora for identification and translation. The conclusion emphasizes the potential of neural machine translation, yet acknowledges challenges in diverse languages, low-resource languages, and specialized terminology.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230482

    Design and development of 2D game 'Adventurous spirit'

    In the present world, 2D games continue to be a significant component of the gaming industry. This article aims to discuss the process of designing and developing the 2D game "Adventurous spirit," while analyzing its essential elements and Unreal Engine technology. We will introduce the principles of game design, the intentions and purposes behind designing the game, the development process, as well as the tools and techniques employed. This 2D game is developed using the currently popular Unreal Engine, showcasing the final game outcome. Through the presentation in this report, readers will gain insights into the fundamental knowledge and practical experiences of 2D game design and development, along with some technical aspects and operations related to Unreal Engine development. Additionally, the genre of this game is a 2D side-scrolling game, where players enhance character abilities and levels through defeating monsters, and they can explore by selecting different characters.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230483

    Strategy optimization for breakout games based on improved BFS algorithm and image recognition techniques

    This paper develops a Dungeon Game and details the development process of the City Game. The Dungeon Game is an adventure-themed role-playing game in which the player takes on the role of a character he or she creates and engages in a variety of adventurous activities in a set fictional world. It also allows the use of various props to add interest and realism to the game. In the development of this game, the pathfinding algorithm is crucial to the character's movement and behavior, and the previous traditional pathfinding method can no longer meet the needs of the game, so we optimize the original algorithm, adding dynamic programming and backtracking method in the original algorithm to improve the original algorithm's pathfinding logic to make the character pathfinding in the game more reasonable, to improve the game's performance and user experience. performance and user experience. At the same time, this paper tries to incorporate an image recognition technique based on ANN neural network into the game by adding figures as images into the game, to enrich the content of the game and make it have more possibilities, which increases the playability for the players. In the path optimization test, although the response time is slightly inaccurate, the overall response time is gradually decreasing as the optimization algorithms improve. These algorithm improvements have a more significant increase in the efficiency of the game.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230484

    Current study on NeRF-based techniques in SLAM frameworks

    This paper comprehensively examines the application and current research status of Neural Radiance Fields (NeRF) technology within Simultaneous Localization and Mapping (SLAM) systems. NeRF, which has gained significant attention since 2020, has evolved into a powerful method for reconstructing 3D scenes from images, offering advantages such as continuous scene representation and photorealistic novel view synthesis. However, it also comes with drawbacks, including substantial training data requirements, limited model generalization, and challenges in map scalability. In contrast, SLAM is a complex, real-time, efficient, and robust system capable of tracking camera motion and constructing environmental maps in real-time, with no limitations on map size. The integration of NeRF technology into SLAM enhances the capabilities of various modules, including Mapping, Tracking, Optimization, Loop Closure, and Localization, providing potential advantages. Beginning with an exploration of NeRF’s fundamental principles and its inherent strengths and weaknesses, this paper delves into the significant implications of integrating NeRF into the SLAM pipeline. It addresses the challenges encountered during implementation and outlines potential future directions. The aim is to provide a clear elucidation of the evolving landscape of the combined NeRF and SLAM approach, serving as a reference for researchers interested in pursuing this research direction.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230486

    A research of artificial intelligence game agent application

    Currently, large language models are on the rise with breakthrough progress in artificial intelligence. Existing reviews of AI game agents have not covered these latest developments, requiring a combing and analysis of the newest research advancements in game AI agents. This paper summarizes the application scenarios of game AI agents in four aspects: combat AI, Non-Player Character (NPC) interaction, automated testing, and Artificial General Intelligence (AGI) testing. In combat AI, there is a progressive developmental trend, with the introduction of Monte Carlo tree search and reinforcement learning enabling AI game agents to fully surpass humans in traditional board games. In NPC interaction, full AI is unnecessary. Game developers only need to incorporate AI for abilities related to player experience to increase appeal, with controllable generation results. In automated testing, game AI agents lack generalizability for testing so far. In AGI testing, academia has helpfully explored general game AI, but capabilities remain limited to certain games. Introducing large language models to game AI agents shows unprecedented capabilities. Finally, this paper provides an outlook on the hot topics and future directions of this research subject.

  • Open Access | Article 2024-02-07 Doi: 10.54254/2755-2721/37/20230489

    The performance of artificial intelligence in medical field

    In modern society, there is an increasing prevalence of individuals who are able to avail themselves of medical facilities for the purpose of receiving healthcare services. With the rise in population, there has been a corresponding decrease in the availability of medical resources for some disorders, leading to challenges in accurately diagnosing patients by healthcare professionals. According to specialists, artificial intelligence (AI) is being considered as a potential solution for addressing medical challenges. This paper mainly focuses on discussing the impact of artificial intelligence in the medical field. Through methods of literature review and analysis, this study explores the fundamental idea of AI and its use in the medical field. Besides, the paper also introduces the potential flaws behind AI in the medical field, and how will artificial intelligence help us further in the medical field. The study reveals that artificial intelligence is extensively employed within the medical sector. Through extensive training, AI has the potential to attain a considerable level of accuracy when it comes to diagnosing various ailments. The level of precision exhibited is akin to that observed in medical practitioners’ diagnoses. Nevertheless, artificial intelligence possesses certain limitations. For instance, in the context of privacy preservation, several patients exhibit a reluctance to divulge their symptoms. In the absence of adequate safeguards for information security, this situation might potentially lead to adverse consequences in the lives of these patients and their interpersonal interactions. Ultimately, AI continues to possess significant potential for advancement within the realm of medicine.

Copyright © 2023 EWA Publishing. Unless Otherwise Stated