Applied and Computational Engineering

- The Open Access Proceedings Series for Conferences

Volume Info.

  • Title

    Proceedings of the 6th International Conference on Computing and Data Science

    Conference Date

    2024-09-12

    Website

    https://www.confcds.org/

    Notes

     

    ISBN

    978-1-83558-457-6 (Print)

    978-1-83558-458-3 (Online)

    Published Date

    2024-06-24

    Editors

    Alan Wang, University of Auckland

    Roman Bauer, University of Surrey

Articles

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241385

    Optimization study of pruning strategy based on KNN trajectory similarity query

    With the development of GPS positioning technology, a large amount of spatiotemporal trajectory data has been generated. Due to the complex structure of trajectory data, which has irregular spatial shapes and continuous temporal sequence attributes, querying massive trajectory data poses certain challenges. The pruning strategy of existing trajectory similarity query methods is not very effective. Even after pruning operations, there are still a large number of trajectories that need to undergo distance calculations to confirm whether they are similar trajectories. This paper proposes multiple local pruning optimization schemes, which maximally reduce the number of trajectories in the candidate set. Specifically, it starts with region pruning in the index space, eliminating index spaces with distances greater than the maximum similarity distance from the query trajectory. Then, it performs distance pruning between trajectories, removing trajectories that do not meet the distance conditions. Finally, it adds a termination condition: when the distance between the current index space and the query trajectory is greater than the maximum similarity distance and the number of trajectories in the result set is K, the loop is exited, and the query process ends. Tests on real datasets demonstrate that the KTSS method outperforms current algorithms of the same type.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241388

    Enhancing conversational recommendation systems through the integration of KNN with ConLinUCB contextual bandits

    In recommender system research, contextual multi-armed bandits have shown promise in delivering tailored recommendations by utilizing contextual data. However, their effectiveness is often curtailed by the cold start problem, arising from the lack of initial user data. This necessitates extensive exploration to ascertain user preferences, consequently impeding the speed of learning. The advent of conversational recommendation systems offers a solution. Through these systems, the conversational contextual bandit algorithm swiftly learns user preferences for specific key-terms via interactive dialogues, thereby enhancing the learning pace. Despite these advancements, there are limitations in current methodologies. A primary issue is the suboptimal integration of data from key-term-centric dialogues and arm-level recommendations, which could otherwise expedite the learning process. Another crucial aspect is the strategic suggestion of exploratory key phrases. These phrases are essential in quickly uncovering users’ potential interests in various domains, thus accelerating the convergence of accurate user preference models. Addressing these challenges, the ConLinUCB framework emerges as a groundbreaking solution. It ingeniously combines feedback from both arm-level and key-term-level interactions, significantly optimizing the learning trajectory. Building upon this, the framework integrates a K-nearest neighbour (KNN) approach to refine key-term selection and arm recommendations. This integration hinges on the similarity of user preferences, further hastening the convergence of the parameter vectors.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241389

    Comparative analysis and applications of classic multi-armed bandit algorithms and their variants

    The multi-armed bandit problem, a pivotal aspect of Reinforcement Learning (RL), presents a classic dilemma in sequential decision-making, balancing exploration with exploitation. Renowned bandit algorithms like Explore-Then-Commit, Epsilon-Greedy, SoftMax, Upper Confidence Bound (UCB), and Thompson Sampling have demonstrated efficacy in addressing this issue. Nevertheless, each algorithm exhibits unique strengths and weaknesses, necessitating a detailed comparative evaluation. This paper executes a series of implementations of various established bandit algorithms and their derivatives, aiming to assess their stability and efficacy. The study engages in empirical analysis utilizing a real dataset, generating charts and data for a thorough examination of the pros and cons associated with each algorithm. A significant aspect of the research focuses on the parameter sensitivity of these algorithms and the impact of parameter tuning on their performance. Findings reveal that the SoftMax algorithm's effectiveness is markedly influenced by the initial estimated mean reward value for each arm. Conversely, algorithms like Epsilon-Greedy and UCB exhibit enhanced performance with optimal parameter settings. Furthermore, the study explores the limitations inherent in classic bandit algorithms and introduces some innovative models and methodologies pertinent to the multi-armed bandit problem, along with their applications.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241390

    Strategic insights from multi-armed bandits: Applications in real-time strategy games

    In real-time strategy games, players often face uncertainty regarding which strategy will lead to victory. This paper delves into how multi-armed bandit (MAB) algorithms can assist in this context, beginning with an exploration of MAB's theoretical principles, particularly the crucial balance between exploration and exploitation. The study compares the efficacy of the Explore-Then-Commit (ETC), Upper Confidence Bound (UCB), and Thompson Sampling (TS) algorithms through practical experimentation. Beyond gaming, the paper also considers the broader implications of MAB algorithms in healthcare, finance, and dynamic pricing within online retail sectors. A focal point of the research is the application of UCB and TS algorithms in StarCraft, a popular real-time strategy game. The performance of these algorithms is rigorously evaluated by calculating the cumulative regret value, a key metric in assessing strategic effectiveness. The findings suggest that the implementation of UCB and TS algorithms significantly enhances players' winning rates in the game. While the results are promising, the paper acknowledges ongoing challenges and encourages further exploration into this fascinating and valuable area of study. This research not only contributes to the understanding of strategic decision-making in gaming but also signals potential cross-sectoral applications of MAB algorithms.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241401

    Optimizing video click-through rates with bandit algorithms

    In recent years, videos have increasingly influenced public perception, making video platforms a focal point of digital consumption. One critical challenge for platform operators is identifying videos that resonate most with users, as user ratings directly reflect viewer preferences and experiences. This study explores the use of bandit algorithms to predict and strategize the overall ratings of various anime videos on the Bilibili platform. Bandit algorithms, a subset of the multi-armed bandit model, dynamically adjust selection strategies based on prior feedback to maximize cumulative rewards. Our empirical research assessed multiple gambling algorithms, including the ε-greedy, Upper Confidence Bound (UCB), Explore-then-Commit (ETC), and Thompson Sampling (TS) algorithms. The findings indicate that the Thompson Sampling algorithm, in particular, achieved the lowest cumulative regret in selecting optimal videos on the Bilibili platform, showcasing its superior performance. This study highlights the potential of bandit algorithms to enhance video selection processes, ensuring that platforms can effectively cater to user preferences and enhance viewer satisfaction.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241402

    Enhancing movie recommendations through comparative analysis of UCB algorithm variants

    In the digital realm, recommendation systems are pivotal in shaping user experiences on online platforms, tailoring content based on user feedback. A notable algorithm in this domain is the multi-armed bandit algorithm, with the Upper Confidence Bound (UCB) emerging as a classic and effective variant. This paper delves into an array of Upper Confidence Bound algorithm variations, encompassing UCB1, Asymptotically Optimal UCB, UCB-V, and UCB1Tuned. The research harnesses the MovieLens dataset to assess the performance of these algorithms, employing cumulative regret as the primary metric. For l in UCB1 and c in UCB-V, both oversized and undersized parameters will result in negative outcomes. And UCB1Tuned outperforms the other three algorithms in this experiment, since it considers variance and adjusts parameters dynamically. The study demonstrates that setting a appropriate UCB index is crucial for enhancing the performance of the UCB algorithm in recommendation system. It holds significance for both improve recommendation system algorithms and enhance user experience.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241403

    Applying Multi-Armed Bandit algorithms for music recommendations at Spotify

    This study explores the application of multi-armed bandit algorithms in enhancing music recommendation systems, with a focus on Spotify. It delves into the Explore-Then-Commit (ETC), Upper Confidence Bound (UCB), and Thompson Sampling (TS) algorithms, evaluating their efficacy within the Spotify context. The primary objective is to determine which algorithm optimally balances exploration and exploitation to maximize user satisfaction and engagement. The research reveals that the ETC algorithm, with its rigid exploration and exploitation phases, incurs a notably higher regret value. This rigidity can lead to missed opportunities in identifying optimal choices and hinder adaptability to evolving user preferences. Conversely, the UCB and TS algorithms exhibit remarkable adaptability and a flexible balance between exploration and exploitation. This flexibility translates into more personalized and satisfactory user experiences in music recommendations. However, the selection of the most appropriate algorithm should be contingent on the size and characteristics of the specific user dataset, as well as the fine-tuning of algorithm parameters to align with user preferences and behaviors.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241404

    Advancing decision-making strategies through a comprehensive study of Multi-Armed Bandit algorithms and applications

    Multi-Armed Bandit (MAB) strategies play a pivotal role in decision-making algorithms by adeptly managing the exploration-exploitation trade-off in environments characterized by multiple options and constrained resources. This paper delves into the core MAB algorithms, including Explore-Then-Commit (ETC), Thompson Sampling, and Upper Confidence Bound (UCB). It provides a detailed examination of their theoretical underpinnings and their application across diverse sectors such as recommender systems, healthcare, and finance. MAB algorithms are celebrated for their efficiency in optimizing decision outcomes; however, they are not without challenges. Significant issues include managing the complexity of exploration and adapting to non-stationary environments where the dynamics of the available options may change over time. A nuanced understanding of these challenges is crucial for effectively implementing MAB strategies in complex decision-making scenarios. This study not only highlights the versatility and potential of MAB algorithms but also underscores the need for ongoing research to refine these techniques and expand their applicability.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241405

    Performance variance in Multi-Armed Bandits: In-depth analysis of three core algorithms

    In real-time strategy games, players grapple with uncertainty regarding the best strategy for victory. This paper delves into multi-armed bandit (MAB) algorithms as potential solutions. The theoretical foundations of MAB are explored, with a focus on the crucial balance between exploration and exploitation. An experimental comparison of the Explore-Then-Commit (ETC), Upper Confidence Bound (UCB), and Thompson Sampling (TS) algorithms is conducted, showcasing their varied performance. Beyond gaming, the paper also examines the broader applications of MAB algorithms in fields such as healthcare, finance, and dynamic pricing in online retail, highlighting their versatility. A significant portion of the study is dedicated to implementing the UCB and TS algorithms in StarCraft, a popular real-time strategy game. The performance of these algorithms is assessed by calculating cumulative regret values, a metric critical to understanding their effectiveness in decision-making contexts. The results indicate that both UCB and TS algorithms substantially improve players' win rates in StarCraft. However, the study acknowledges existing challenges and the need for further research in this area. The use of MAB algorithms in complex, dynamic environments like real-time strategy games presents a rich avenue for exploration and holds significant promise for enhancing decision-making strategies in diverse domains. This research, therefore, not only contributes to the understanding of MAB algorithms in gaming but also underscores their potential in various other sectors.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241406

    Optimizing decision-making in uncertain environments through analysis of stochastic stationary Multi-Armed Bandit algorithms

    Reinforcement learning traditionally plays a pivotal role in artificial intelligence and various practical applications, focusing on the interaction between an agent and its environment. Within this broad field, the multi-armed bandit (MAB) problem represents a specific subset, characterized by a sequential interaction between a learner and an environment where the agent’s actions do not alter the environment or reward distributions. MABs are prevalent in recommendation systems and advertising and are increasingly applied in sectors like agriculture and adaptive clinical trials. The stochastic stationary bandit problem, a fundamental category of MAB, is the primary focus of this article. Here, we delve into the implementation and analytical comparison of several key bandit algorithms—including Explore-then-Commit (ETC), Upper Confidence Bound (UCB), Thompson Sampling (TS), Epsilon-Greedy (ɛ-Greedy), SoftMax, and Conservative Lower Confidence Bound (CON-LCB)—across various datasets. These datasets vary in the number of options (arms), reward distributions, and specific parameters, offering a broad testing ground. Additionally, this work provides an overview of the diverse applications of bandit problems across different fields, highlighting their versatility and broad impact.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241408

    Analysis the approaches and applications for jazz music composing based on machine learning

    As a matter of fact, computer composing is a hot topic for study in recent years. To be specific, Jazz, one of the essential music genres, has a complex and irregular musical structure. Current researchers are focusing on how to use models to generate expressive and innovative jazz. This study first summarizes in detail the characteristics of the musical structure of jazz and the unique structures of jazz music that are the difficulties of model training. At the same time, it then analyzes the feasibility and application of several popular machine learning models in jazz music composition. Finally, this study combines the current development of the field of computer composition with the challenges faced by the field of computer-generated jazz, as well as the direction of the next step in the development of continued efforts to explore in-depth, in the hope that this will provide researchers with new research perspectives, and to promote the development of new forms of jazz music can flourish in the age of intelligent machines.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241413

    Analysis of dreamcore music composing based on Nyquist

    As a matter of fact, computer music composition serves as a potent tool for musicians in the process of sound synthesis. More of their ideas and imaginations can be realized, more innovative exploration could be fostered in the field of music. Dreamcore is an art style that presents a kind of surreal nostalgia, and dreamcore music is a relatively new musical style related to the feeling of dreamcore. Being attracted by this musical style, the author aimed to compose a piece of dreamcore music by Nyquist, which is an interactive language for music composition. After summarizing and imitating the music styles and characteristics of the selected sample music, this study used the score, code it in Nyquist, and use musical effects to create a piece of music that is dreamy. The research obtained a music demo after the composition. The piece of music did embody the dreamcore aesthetic, and the basic requirements that the author made were met, but further exploration and enhancement could still be done to the music.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241414

    Composing jazz music pieces using LSTM neural networks approach

    The domain of artificial intelligence has increasingly extended into the creative arts, aiming to emulate and augment human creativity with automated processes. This is particularly evident in the field of music, where AI's ability to learn and produce intricate compositions has attracted significant attention. This study explores the challenge of generating jazz music using artificial intelligence, specifically focusing on the application of Long Short-Term Memory (LSTM) neural networks for jazz composition. By optimizing the model to address the genre's complexity, the research demonstrates the LSTM's capacity to capture and reproduce jazz's essential harmonic progressions and rhythmic nuances. Quantitative analyses show high accuracy and a deep understanding of musical structures, whereas qualitative feedback confirms the model's efficacy in producing compositions that embody jazz's spontaneity. Despite its achievements, the model's tendency to generate repetitive sequences suggests areas for improvement. This paper advances the field of AI in music, illustrating the potential of LSTM networks to mimic complex musical genres and emphasizing the necessity of ongoing model refinement to foster creativity. It highlights the evolving role of machine learning in music generation, proposing a foundation for future work aimed at diminishing the gap between AI capabilities and artistic expression.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241415

    Analysis of the realization for emotion recognition based on machine learning

    As a matter of fact, music emotion recognition based on machine learning has been a trend in recent years. This study describes the analysis of the realization for emotion recognition based on machine learning. In this paper, the study expresses some methods of analyzing the emotion of the music composition. The paper represents some classifier molds that enable us to determine the emotion of the music. It aims to provide readers with a deeper comprehension of the music and composers’ feelings and to know more about the background of the music. The study may let more people to improve their own ability to appreciate music, if this process has let great amounts of individuals known. It also allows people to have more awareness and feelings with more voices. In this paper, the research will also show some about the evolution and original of the synthesizer. Moreover, the study will go to illustrate this through scientific, real examples and personal experiences and pictures.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241416

    New York city taxi demand forecast based on ARIMA model

    The competition in the taxi market is becoming increasingly fierce, but there is a gap between the demand and the number of taxis in some time periods, which not only intensifies ineffective competition among taxis but also brings inconvenience to passengers. This study aims to establish a predictive model to predict the demand for taxis in different time periods in the city. The data was collected from New York City yellow taxi data which was from June 1st, 2022 to June 6th, 2022. After processing the raw data, the optimal parameter selection of the model is determined through ADF testing to improve accuracy. Through ACF and PACF calculation, the data and images are analyzed to find the most suitable p and q values. Use ARIMA model to fit the data and obtain a model with robust fitting parameters. The distribution of predicted values is very consistent with actual data. The model was used for 50 periods of prediction, and through analysis of the research results, the fitting effect of the prediction was good. It was found that the accuracy of the model was high, proving its ability to predict short-term taxi demand.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241418

    The evolution and impact of Multi-Armed Bandit algorithms in social media

    This paper examines the transformative impact of Multi-Armed Bandit (MAB) algorithms on user experiences across social media platforms. Initially conceptualized in the 1930s and formalized in the 1950s, MAB algorithms have become foundational to the evolution of digital interactions and content personalization. These algorithms adeptly navigate the trade-off between exploration and exploitation to maximize user engagement and satisfaction. By scrutinizing their implementation from early adopters like Yahoo to contemporary giants such as Facebook, Instagram, and TikTok, this analysis elucidates the algorithms' prowess in tailoring content recommendations, refining advertising strategies, and bolstering overall platform engagement. Moreover, this study addresses the ethical dimensions of MAB algorithms, with a particular emphasis on concerns surrounding user privacy and the perpetuation of echo chambers. Through an extensive synthesis of theoretical insights and empirical applications, this paper highlights the pivotal role of MAB algorithms in shaping the digital and social media landscape, advocating for future research focused on improving algorithmic transparency and ethical governance.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241421

    Analysis of the implementation methods for emotion recognition in music based on machine learning

    Music is a carrier of emotions, capable of triggering, expressing, and conveying people's innermost emotional experiences. With the development of digital technology, the recognition of emotions in music has been the subject of extensive research. From the viewpoint of application, MER is of great significance in music personalized recommendation services, psychotherapy, music creation, and music visualization. Currently, MER implementation methods based on machine learning are gradually becoming mainstream. This study starts from the basic knowledge of MER, first introduces the relevant research directions and research background of MER, and then proposes a three-part research framework combining MER and MEC. Based on this research framework, the machine learning algorithms and specific applications involved in the framework are introduced. Finally, the challenges and future paths facing the implementation of MER technology are proposed. Provide a reference for improving the accuracy and quality of using artificial intelligence methods to capture music characteristics and expressiveness.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241430

    Animal speech and singing synthesis model based on So-VITS-SVC

    Currently, when researchers in deep learning and neural network technology have made significant progress, the author makes a new bold attempt to apply the technical principles of speech and singing synthesis with artificial intelligence to the field of animal speech and singing synthesis, using So-VITS-SVC4.0 framework, which was originally designed for human voice synthesis. Taking dogs as an example of a species and putting datasets of their sounds to use, the author is committed to capturing its sound characteristics and vocalization through model training and generating synthetic sounds with a high degree of similarity. This research may not only contribute to a deeper understanding of how animals communicate, but also open up new possibilities for animal sound art and music creation. With the continuous progress and improvement of technology, synthetic animal speech and singing by artificial intelligence may play an increasingly important role in zoological research and entertainment, bringing new perspectives and possibilities for communication between humans and animals.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241431

    Synthesis of natural sound based on Nyquist

    In recent years, sound synthesis has become an important part of daily life with the rapid development of the technology. There are many methods of natural sound synthesis, but each of them has its disadvantages. This paper explored a novel method of natural sound synthesis based on Nyquist. This method has significant advantages in high speed, convenience, flexibility, randomness, and small file size. The research investigates the effectiveness of this method by synthesizing four distinct natural sounds - rain, wind, thunder, and a combination of all three - using unit generators in Nyquist. To further evaluate the result, Mel spectrograms and a questionnaire are applied. The evaluation shows that the synthesized sounds meet expectations and are relatively real and natural. This method can be combined with other technology, such as the large language model, to provide better accessibility. This method can generate natural sounds for many tasks, especially those requiring high speed.

  • Open Access | Article 2024-06-06 Doi: 10.54254/2755-2721/68/20241432

    Analyzing principles and applications of machine learning in music: Emotion music generation, and style modeling

    The combination of machine learning with music composition and production is proving viable for innovative applications, enabling the creation of novel musical experiences that were once the exclusive domain of human composers. This paper explores the transformative role of machine learning in music, particularly focusing on emotion music generation and style modeling. Through the development and application of models including DNNs, GANs, and Autoencoders, this study delves into how machine learning is being harnessed to not only generate music that embodies specific emotional contexts but also to transfer distinct musical styles onto new compositions. This research discusses the principles of these models, their operational mechanisms, and evaluates their effectiveness through various metrics such as accuracy, precision, and creative authenticity. The outcomes illustrate that these technologies not only enhance the creative possibilities in music but also democratize music production, making it more accessible to non-experts. The implications of these advancements suggest a significant shift in the music industry, where machine learning could become a central component of creative processes. These results pave a path to the understanding of the potential and limitations of machine learning in music and forecasts future trends in this evolving landscape.

Copyright © 2023 EWA Publishing. Unless Otherwise Stated