Article
Computer Science, Artificial Intelligence
Alessandra De Paola, Salvatore Gaglio, Andrea Giammanco, Giuseppe Lo Re, Marco Morana
Summary: Modern smart environments present challenges in designing intelligent algorithms to assist users, such as trajectory recommendations and itinerary planning in the face of diverse points of interest. A multi-agent itinerary suggestion system is proposed to address these challenges, utilizing reinforcement learning to provide high-quality suggestions and overcome issues like cold-start and preference elicitation. Real-life deployments have shown the effectiveness of the approach in scenarios such as smart campuses and theme parks.
CAAI TRANSACTIONS ON INTELLIGENCE TECHNOLOGY
(2021)
Article
Chemistry, Analytical
Saif Ur Rehman, Noha Alnazzawi, Jawad Ashraf, Javed Iqbal, Shafiullah Khan
Summary: Internet of Things (IoT)-backed smart shopping carts generate data that can be used for business goal and strategy setting through AI and data mining techniques. The challenge lies in extracting the top-K most frequent patterns, for which an efficient algorithm, TKIFIs Miner, is proposed.
Article
Engineering, Civil
Tiantian Yang, Lujun Zhang, Taereem Kim, Yang Hong, Di Zhang, Qidong Peng
Summary: This study compares 12 different parameterized AI&DM models in simulating controlled reservoir outflows in the Upper Colorado Region, United States, and finds that the Random Forecast and Long-Short-Term-Memory models consistently exhibit the best statistical performance. The Multi-Layer Perceptron model and Extreme Gradient Boosting Tree Algorithm perform more stably under complex input scenarios. The performance of different AI&DM models is closely related to the characteristics of the reservoirs.
JOURNAL OF HYDROLOGY
(2021)
Article
Environmental Sciences
Francesca Larosa, Sergio Hoyas, Javier Garcia-Martinez, J. Alberto Conejero, Francesco Fuso Nerini, Ricardo Vinuesa
Summary: Large language models provide an opportunity to advance climate and sustainability research. We believe that regulating and validating generative artificial intelligence models would benefit society more than stopping development.
NATURE CLIMATE CHANGE
(2023)
Article
Automation & Control Systems
Yuming Li, Wei Zhang, Yanyan Liu, Rudong Jing, Changsong Liu
Summary: This paper proposes an object detection model based on DETR for fire and smoke detection, which simplifies the detection pipeline and builds an end-to-end detector. By adding a normalization-based attention module in the feature extraction stage and using multiscale deformable attention in the encoder-decoder structure, the model achieves improved detection performance while reducing complexity.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Lidong Wang, Fang Zhao, Hongyue Guo, Xiaodong Liu, Witold Pedrycz
Summary: This article introduces a top-down granulation model adhering to the principle of justifiable granularity, incorporating principal component analysis to capture the correlation among features. Experimental results suggest the feasibility of the models and illustrate the impact of parameter values on the constructed information granules.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Chemistry, Multidisciplinary
Wendy L. Williams, Lingyu Zeng, Tobias Gensch, Matthew S. Sigman, Abigail G. Doyle, Eric Anslyn
Summary: Organic chemistry involves complex relationships such as the structure of a reactant and the resulting product, reaction conditions and yield, and the structure of a catalyst and enantioselectivity. Data-driven modeling is an established and contemporary approach to exploring these questions, with case studies demonstrating how the community has advanced physical organic chemistry tools with the help of computers and data to enhance expert chemists' intuition and facilitate the prediction of structure-activity and structure-property relationships.
ACS CENTRAL SCIENCE
(2021)
Article
Computer Science, Information Systems
Silvio Barra, Salvatore M. Carta, Alessandro Giuliani, Alessia Pisu, Alessandro Sebastian Podda, Daniele Riboni
Summary: This article introduces an AI-based system for football match annotation, which utilizes a mixed user interface to annotate football matches and processes players' motor performance using machine learning algorithms. The experimental results of the system demonstrate its effectiveness in real-world adoption scenarios.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Construction & Building Technology
M. Z. Naser
Summary: The research demonstrates the importance of utilizing modern computing techniques, such as data science and machine learning algorithms, in structural fire engineering applications for analyzing and predicting fire-induced spalling phenomenon.
JOURNAL OF MATERIALS IN CIVIL ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Razieh Davashi
Summary: In this paper, a fast method called ITUFP is proposed for interactive mining of Top-K UFPs. The method efficiently stores and extracts pattern information by creating UP-Lists and IMCUP-Lists, and only updates the IMCUP-Lists when the K value changes. Experimental results demonstrate that the proposed method is very efficient for interactive mining of Top-K UFPs.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Xin Wang, Zhuo Lan, Yu-Ang He, Yang Wang, Zhi-Gui Liu, Wen-Bo Xie
Summary: This article introduces a cost-effective approach for frequent pattern mining on large graphs. The approach applies a level-wise strategy to incrementally detect frequent patterns and can terminate the mining process once the top-k patterns are discovered. It also utilizes a smart traverse strategy and compact data structures to compute the lower bound of support.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Matteo Cucchi, Christopher Gruener, Lautaro Petrauskas, Peter Steiner, Hsin Tseng, Axel Fischer, Bogdan Penkovsky, Christian Matthus, Peter Birkholz, Hans Kleemann, Karl Leo
Summary: Early detection of malign patterns in patients' biological signals is crucial, and organic electrochemical devices are considered ideal for biosignal monitoring. The study demonstrates the potential of brain-inspired networks composed of organic electrochemical transistors for time-series predictions and classification tasks, showing promise for biofluid monitoring and biosignal analysis.
Article
Engineering, Biomedical
Carmen Camara, Pedro Peris-Lopez, Masoumeh Safkhani, Nasour Bagheri
Summary: The study introduces an innovative identification technique using electrocardiograms and musical features, achieving high accuracy in identity verification by converting ECGs into audio files and extracting features.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2022)
Article
Computer Science, Information Systems
Ham Nguyen, Tuong Le
Summary: This study presents a robust method for mining top-rank-k erasable closed patterns (ECPs) and combines the mining and ranking phases into a single step to improve efficiency. Experimental results confirm that this method outperforms other approaches in mining top-rank-k ECPs.
CMC-COMPUTERS MATERIALS & CONTINUA
(2022)
Article
Computer Science, Information Systems
Chunkai Zhang, Zilin Du, Wensheng Gan, Philip S. Yu
Summary: High-utility sequential pattern mining (HUSPM) has attracted significant research interest recently, with the main task of finding subsequences with high utility in a quantitative sequential database. The top-k HUSPM concept was introduced to address the challenge of specifying a minimum utility threshold. Existing strategies for top-k HUSPM require improvement in terms of efficiency and scalability.
INFORMATION SCIENCES
(2021)