Large-scale Sets Research Articles

Quasi-Newton Acceleration of EM and MM Algorithms via Broyden’s Method

The principle of majorization-minimization (MM) provides a general framework for eliciting effective algorithms to solve optimization problems. However, the resulting methods often suffer from slow convergence, especially in large-scale and high-dimensional data settings. This has motivated several acceleration schemes tailored for MM algorithms, but many existing approaches are either problem-specific, or rely on approximations and heuristics loosely inspired by the optimization literature. We propose a novel quasi-Newton method for accelerating any valid MM algorithm, cast as seeking a fixed point of the MM algorithm map. The method does not require specific information or computation from the objective function or its gradient, and enjoys a limited-memory variant amenable to efficient computation in high-dimensional settings. By rigorously connecting our approach to Broyden’s classical root-finding methods, we establish convergence guarantees and identify conditions for linear and super-linear convergence. These results are validated numerically and compared to peer methods in a thorough empirical study, showing that it achieves state-of-the-art performance across a diverse range of problems. Supplementary materials for this article are available online.

DLSDHMS: Design of a deep learning-based analysis model for secure and distributed hospital management using context-aware sidechains

Designing an efficient hospital management solution requires the integration of multidomain operations that include secure storage, alert system modelling, infrastructure management, staff management, report analysis, and feedback-based learning tasks. Existing hospital management models are either highly complex or do not incorporate comprehensive deep learning analysis, which limits their deployment capabilities. Moreover, most of these models use mutable storage solutions, which restricts their trust levels under multi-patient to multi-doctor mapping scenarios. To overcome these issues, this article proposes the design of a novel deep Learning-based analysis model for secure and distributed hospital management via context-aware sidechains. The model initially collects large-scale information sets from different hospital entities via an IoT-based network and stores the information on context-sensitive sidechains. These context-sensitive sidechains store information sets related to Medicine Management, Doctor Management, Insurance and Billing Management, and Appointment Management operations. These chains are optimized via an Iterative Genetic Algorithm (IGA) that assists in improving storage and retrieval performance via intelligent merging and splitting operations. Information stored on these chains is processed via a combination of Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), that assist in identifying patient-level diseases and issues. The information obtained from these classifiers is updated on the central repository and assists in the pre-emption of diseases for other patients. Due to these integrations, the proposed model is capable of reducing computational delay by 3.5 % and reducing storage cost by 8.3 % when compared to other blockchain-based deployments. The model is also able to pre-empt patient issues with 9.3 % higher accuracy and 4.8 % higher precision, which makes it useful for real-time clinical deployments.

Monitoring and Mapping Urban Sprawl Over Heritage Hotspots Using Copernicus Land Monitoring Services: The case of periurban large-scale, wind-powered water extraction mills in Palma (Mallorca)

Among the varied group of human constructions of heritage interest are the old wind-powered water extraction mills that sometimes form large-scale sets in rural areas with high wind potential. On the Mediterranean island of Mallorca (Spain), up to 2,400 windmills dating from the 19th and early 20th centuries have been counted. The areas in which they are located are considered heritage ‘hotspots’, i.e., areas prone to specific problems, such as the progressive abandonment of ethnological heritage resulting from urban sprawl over areas with an agricultural orientation. This article aims to monitor urban sprawl in the municipality of Palma, to quantify and map its impact on a set of windmills located mainly in the plain of Sant Jordi, to the east of the city. The study has been carried out using methodologies and analysis techniques from the Copernicus Land Monitoring Service’s Urban Atlas and Imperviousness Density products. The study shows that areas with agricultural land uses have been progressively transformed into urbanised ones. This transformation has impacted, above all, windmills located in peri-urban areas adjacent to the city. The analysis aims to show the analytical possibilities of Copernicus services and products, and their applicability in the planning and management of peri-urban agro-industrial heritage.

Improved Gravity Inversion Method Based on Deep Learning with Physical Constraint and Its Application to the Airborne Gravity Data in East Antarctica

This paper aims to solve the limitations of traditional gravity physical property inversion methods such as insufficient depth resolution and difficulties in parameter selection, by proposing an improved 3D gravity inversion method based on deep learning. The deep learning network model is established using the fully convolutional U-net network. To enhance the generalization ability of the sample set, the large-scale training set and test set are generated by the random walk, based on the forward theory. Founded on the traditional loss function’s definition, this paper introduces an improvement incorporating a physical constraint to measure the degree of data fitting between the predicted and the real gravity data. This improvement significantly boosted the accuracy of the deep learning inversion method, as verified through both a single model and an intricate combination model. Finally, we applied this improved inversion method to the gravity data from the Gamburtsev Subglacial Mountains in the interior of East Antarctica, obtaining a comprehensive 3D crustal density structure. The results provide new evidence for the presence of a dense crustal root situated beneath the central Gamburtsev Province near the Gamburtsev Suture.

Developing a digital tutor as an intermediary between students, teaching assistants, and lecturers

Individualized learning support is an essential part of formal educational learning processes. However, in typical large-scale educational settings, resource constraints result in limited interaction among students, teaching assistants, and lecturers. Due to this, learning success in those settings may suffer. Inspired by current technological advances, we transfer the concept of chatbots to formal educational settings to support not only a single task but a full lecture period. Grounded on an expert workshop and prior research, we design a natural language-based digital tutor acting as an intermediary among students, teaching assistants, and lecturers. The aim of the digital tutor is to support learners automated during the lecture period in natural language-based chat conversations. We implement a digital tutor in an iterative design process and evaluate it extensively in a large-scale field setting. The results demonstrate the applicability and beneficial support of introducing digital tutors as intermediaries in formal education. Our study proposes the concept of using digital tutors as intermediaries and documents the development and underlying principles.

MPFFNet: LULC classification model for high-resolution remote sensing images with multi-path feature fusion

ABSTRACT Land Use/Land Cover (LULC) classification has become increasingly important in various fields, including ecological and environmental protection, urban planning, and geological disaster monitoring. With the development of high-resolution remote sensing satellite technology, there is a growing focus on achieving precise LULC classification. However, the accuracy of fine-grained LULC classification is challenged by the high intra-class diversity and low inter-class separability inherent in high-resolution remote sensing images. To address this challenge, this paper proposes a novel multi-path feature fusion semantic segmentation model, called MPFFNet, which combines the segmentation results of convolutional neural networks with traditional filtering processes to achieve finer LULC classification. MPFFNet consists of three modules: the Improved Encoder Module (IEM) extracts contextual and spatial detail information through the backbone network, DASPP, and MFEAM; the Improved Decoder Module (IDM) utilizes the Cascade Feature Fusion (CFF) module to effectively merge shallow and deep information; and the Feature Fusion Module (FAM) enables dual-path feature fusion using a convolutional neural network and Gabor Filter. Experimental results on the large-scale classification set and the fine land-cover classification set of the Gaofen Image Dataset (GID) demonstrate the effectiveness of the proposed method, achieving mIoU scores of 81.02% and 77.83%, respectively. These scores outperform U-Net by 7.95% and 3.28%, respectively. Therefore, we believe that our model can deliver superior results in the task of LULC classification.

Cross-Region Courier Displacement for On-Demand Delivery With Multi-Agent Reinforcement Learning

On-demand delivery has become prevailing for people to order meals and groceries online, especially during the pandemic. It is essential to dispatch massive orders to limited couriers to satisfy on-demand delivery users, especially during peak hours. Existing studies mainly focus on order dispatching within a region, and they are challenging to be applied to the cross-region courier displacement problem due to (i) unique practical factors, including regional spatial-temporal demand-supply dynamics and strict delivery time constraints, and (ii) the large-scale setting and high-dimensional decision space given massive couriers in on-demand delivery. To address these challenges, in this work, we propose an efficient cross-region courier displacement framework, i.e., Courier Displacement Reinforcement Learning (short for CDRL) based on centralized multi-agent actor-critic, which first design the actor-critic network with a time-varying displacement intensity control module to capture demand-supply dynamics and utilize the centralized training and decentralized execution multi-agent framework to address the large-scale coordination. One-month real-world order records collected from one of the biggest on-demand delivery services in the world are utilized to show the performance of our design. The extensive results show that our method offers a 47.97% of increase in balancing supply and demand and reduces idle ride time by 24.62% simultaneously.

A-355 Eyescore, A Point of Care Smartphone App for Early Dry Eye Diagnosis and Management

Abstract Current diagnosis of dry eye disease (DED) has significant challenges with limited accuracy and poor correlation of clinical symptoms. The goal of this study is to develop a smartphone App, namely “EyeScore”, to provide a point of care (POC) and digital solution for early diagnosis of DED. The authors tested the hypothesis of measuring eye blink rate and partial blink count as early clinical biomarkers for the calculation of so called “eye healthiness score”, which allows a convenient, rapid eye exam at home with low resource settings. EyeScore uses an iPhone as an imaging and sensing device for in-App recordings of eyelid movements. The use of facial landmark recognition and eye aspect ratio (EAR) enabled comprehensive digital analysis of video frames for determination of the eye opening/closed states. The smartphone videos from 10 DED patients and 10 healthy controls were tested to optimize EAR derived thresholds for accurate measurements of blink rates and partial blink counts. The authors formulated a clinically relevant algorithm for the calculation of “eye healthiness score”. This 10-point scale score can be easily measured anytime with a non-invasive manner and remotely shared with the patient's eye doctor. As a result, both patients and doctors can monitor the eye conditions over time. Our results showed that EyeScore confirmed the diagnosis of all 10 DED patients. Importantly, Eyescore also identified three individuals with “pre-DED” conditions from 10 healthy controls, demonstrating its potential clinical values for early diagnosis and management of DED. Materials and Methods An Apple MacBook Pro installed with iOS software (iOS 16.1.2) and Xcode 14.0 was used for coding and debugging. Visionkit SDK from Apple Developer was adopted to extract coordinate data from six eye facial landmarks. EAR values were calculated based on the previously published formula, with a dynamic EAR developed as a threshold for the measurement of blink rate and half blink count. For eye healthiness score, a 10-point formula was formulated (0–3, normal; 4–5, Pre-DED; 6–10, DED) with several contributing factors including blink rate, half blink count, eye discomfort and demographic characteristics. Results A total of 20 iPhone videos from DED patients and healthy controls were analyzed and evaluated with EyeScore App. All DED patients obtained scores above 7, confirming their DED diagnosis. On the other hand, three controls obtained scores of 4 or 5, suggesting they were in the Pre-DED state with no noticeable clinical DED symptoms. All other controls obtain scores blow 3 with normal eye conditions. All tests were performed with the consents from the test participants. Conclusions Smart phone-based platform offers great potentials for point of care diagnostic methods. A POC enabled EyeScore App has been successfully developed for early diagnosis of DED. In the future, the authors plan to test EyeScore with a large-scale sample set of DED patients and controls to further validate the accuracy and predictability of eye healthiness score for its broader clinical applications.

Predicting and Visualizing Human Soundscape Perception in Large-Scale Urban Green Spaces: A Case Study of the Chengdu Outer Ring Ecological Zone

Human soundscape perceptions exist through the perceived environment rather than the physical environment itself and are determined not only by the acoustic environment but also by the visual environment and their interaction. However, these relationships have mainly been established at the individual level, which may impede the efficient delivery of human-oriented considerations in improving the quality of large-scale urban spaces. Using the Chengdu Outer Ring Ecological Zone as an example, this study aims to develop an approach to predict human perceptions in large-scale urban green spaces. The site’s visual attributes, i.e., landscape composition, were calculated using space syntax and the quantum geographic information system (QGIS); its aural attributes, i.e., the sound level, were measured on site using a multi-channel signal analyzer; and its functional attributes, i.e., vitality, were documented through on-site observations and mapping. This was performed whilst obtaining people’s perceived soundscape through sound walks and a questionnaire-based on-site survey. The above environmental information was collected at micro-scale measurement spots selected within the site and then used together to formulate a model for predicting people’s soundscape perceptions in the whole site. The prediction results suggested that people’s perceived soundscape satisfaction increased as the distance from the ring road increased, and it gradually reached its highest level in the green spaces stretched outside the ring road. The prediction results of soundscape perception were then visualized using QGIS to develop planning and design implications, along with maps describing the site’s visual, aural, and functional features. Planning and design implications were suggested, including setting green buffers between noise sources and vulnerable areas; identifying and preserving areas with special visual and acoustic characteristics; employing sound shields around traffic facilities; and using natural landscapes to distract people’s attention from noise and to block their view of the source of noise. This study innovatively predicts individual-scale soundscape perception in large-scale UGSs based on environmental visual, aural, and functional characteristics through cross-level measurements, analyses, and model construction. By introducing a systematic perspective, the outcome of this study makes people’s soundscape perceptions more applicable in the planning and design practices of large-scale urban settings.

Nanoscale Titanium Surface Engineering via Low-Temperature Hydrothermal Etching for Enhanced Antimicrobial Properties.

Bioinspired nanotopography artificially fabricated on titanium surfaces offers a solution for the rising issue of postoperative infections within orthopedics. On a small scale, hydrothermal etching has proven to deliver an effective antimicrobial nanospike surface. However, translation to an industrial setting is limited by the elevated synthesis temperature (150 °C) and associated equipment requirements. Here, for the first time, we fabricate surface nanostructures using comparatively milder synthesis temperatures (75 °C), which deliver physicochemical properties and antimicrobial capability comparable to the high-temperature surface. Using a KOH etchant, the simultaneous formation of titania and titanate crystals at both temperatures produces a one-dimensional nanostructure array. Analysis indicated that the formation mechanism comprises dissolution and reprecipitation processes, identifying the deposited titanates as hydrated layered tetra-titanates (K2Ti4O9·nH2O). A proposed nanospike formation mechanism was confirmed through the identification of a core and outer shell for individual nanostructures, primarily comprised of titanates and titania, respectively. Etching conditions dictated crystalline formation, favoring a thicker titanate core for nanorods under higher synthesis temperatures and etchant concentrations. A bactericidal investigation showed the efficacy against Gram-negative bacteria for a representative low-temperature nanosurface (34.4 ± 14.4%) was comparable to the higher temperature nanosurface (34.0 ± 17.0%), illustrating the potential of low-temperature hydrothermal synthesis. Our results provide valuable insight into the applicability of low-temperature etching protocols that are more favorable in large-scale manufacturing settings.

Провозглашение Десятилетия как инструмента стратегического планирования социально-экономического развития регионов (на примере Северо-Западного федерального округа)

The article discusses the announcement of the Decade of Science and Technology as a new strategic planning tool. It is stated that this tool has the character of a long-term large-scale set of measures based on a program-targeted approach. Drawing up regional plans for the implementation of the Decade’s initiatives, according to the Decree of the President of the Russian Federation of 25 April 2022, 231 is a recommended norm for the regions. In this regard, not all regions (in particular, in the North-Western Federal District 6 out of 11) have nor-matively become participants of the Decade. Conclusion dwells upon the fact that it is necessary to expand the participation of regions and municipalities in the implementation of the Decade of Science and Technology, including through the creation of a model plan for the implementation of measures.

Exploring the ability of machine learning-based virtual screening models to identify the functional groups responsible for binding

Many recently proposed structure-based virtual screening models appear to be able to accurately distinguish high affinity binders from non-binders. However, several recent studies have shown that they often do so by exploiting ligand-specific biases in the dataset, rather than identifying favourable intermolecular interactions in the input protein-ligand complex. In this work we propose a novel approach for assessing the extent to which machine learning-based virtual screening models are able to identify the functional groups responsible for binding. To sidestep the difficulty in establishing the ground truth importance of each atom of a large scale set of protein-ligand complexes, we propose a protocol for generating synthetic data. Each ligand in the dataset is surrounded by a randomly sampled point cloud of pharmacophores, and the label assigned to the synthetic protein-ligand complex is determined by a 3-dimensional deterministic binding rule. This allows us to precisely quantify the ground truth importance of each atom and compare it to the model generated attributions. Using our generated datasets, we demonstrate that a recently proposed deep learning-based virtual screening model, PointVS, identified the most important functional groups with 39% more efficiency than a fingerprint-based random forest, suggesting that it would generalise more effectively to new examples. In addition, we found that ligand-specific biases, such as those present in widely used virtual screening datasets, substantially impaired the ability of all ML models to identify the most important functional groups. We have made our synthetic data generation framework available to facilitate the benchmarking of new virtual screening models. Code is available at https://github.com/tomhadfield95/synthVS.

Distance-based generation of a unicursal random path on a non-grid point set for optical polishing.

In modern ultra-precision polishing, sub-aperture technologies are prone to mid-spatial frequency errors due to identical patterns of a path. A random tool path on a regular point set is widely used to suppress mid-spatial frequency errors. In this study, two non-grid uniform point sets, the Fibonacci and the three-directional, were introduced into optical polishing. To solve the time-consuming problem caused by a large amount of distance calculation, a distance-based weighted random (DBWR) algorithm and a linear programming and connecting (LPC) algorithm were presented. The DBWR algorithm reduces the generation time by strengthening the weight of the neighboring points in a specific direction, while the LPC algorithm adjusts the order and distance of points artificially. Then a random stitching method was proposed for the large-scale point set applying to large-sized optical surfaces, which dramatically reduced the generation time. Finally, experiments validated that the algorithms for non-grid sets can be effectively used for optical surface figuring without introducing an apparent mid-spatial frequency.

ARTIFICIAL INTELLIGENCE AND BLOCKCHAIN TECHNOLOGY FOR SECURE DATA AND PRIVACY

Data, which is then processed in order to extract the desirable characteristics, serves as the input for a number of different AI algorithms. However, the facts that can be found on the internet are incredible and tough to authenticate. Given the complexity of the internet, it is quite challenging to validate the data for the consumers. Consequently, in this research, we suggested using SecNet as a solution. An architecture that assists in the protection of data storage, the processing of data, and the sharing of large-scale Internet settings is known as SecNet. The primary objective of this architecture is to enhance the performance of artificial intelligence algorithms across a variety of data sources in order to provide a cyberspace that is more safe and to make use of actual big data. This architecture combines and supplies the following three primary components: 1) The trading of data based on a blockchain is carried out with the ownership of the data being assured. This enables the interchange of accurate data in a wide-ranging environment and contributes to the formation of actual "big data. 2) The protection of an AI-based secure computing platform that is powered by artificial intelligence to develop more astute security standards and contribute to the establishment of a cyberspace that is more reliable. 3) The trustworthy value-sharing Security Service buy Mechanism gives participants a fantastic opportunity to gain Economic Rewards for the provision of data or services, which makes data sharing easier and ultimately leads to improved AI performance".

Nutrient enrichment mediates the effect of biodegradable and conventional microplastics on macroinvertebrate communities

There is growing concern regarding the lack of evidence on the effects bioplastics may have on natural ecosystems, whilst their production continues to increase as they are considered as a greener alternative to conventional plastics. Most research is limited to investigations of the response of individual taxa under laboratory conditions, with few experiments undertaken at the community or ecosystem scale, either investigating microplastics independently or in combination with other pollutants, such as nutrient enrichment. The aim of this study is to experimentally compare the effects of oil-based (high density polyethylene – HDPE) with those of bio-based biodegradable (polylactic acid – PLA) microplastics and their interaction with nutrient enrichment on freshwater macroinvertebrate communities under seminatural conditions. There were no significant differences in total abundance, alpha and beta diversities, or community composition attributable to the type of microplastics, their concentration, or nutrient enrichment compared with the control. However, there was a significant difference in macroinvertebrate alpha diversity between high concentrations of both microplastic types under ambient nutrient conditions, with lower diversity in communities exposed to HDPE compared with PLA. Nutrient enrichment mediated the effect of microplastic type, such that the diversity of macroinvertebrate communities exposed to HDPE were similar to those communities exposed to PLA. These findings suggest that the effects of microplastic pollution on macroinvertebrate communities are very weak at large-scale settings under seminatural conditions and that these effects might be mediated by the nutrient status of freshwater ecosystems. More research under large-scale, long-term, seminatural settings are needed in order to elucidate the impact of both conventional plastics and bioplastics on natural environments and their interactive effect with other occurring stressors and pollutants.

Blind image quality assessment by pairwise ranking image series

Image quality assessment (IQA) is constantly innovating, but there are still three types of stickers that have not been resolved: the "content sticker" - limitation of training set, the "annotation sticker" - subjective instability in opinion scores and the "distortion sticker" - disordered distortion settings. In this paper, a No-Reference Image Quality Assessment (NR IQA) approach is proposed to deal with the problems. For "content sticker", we introduce the idea of pairwise comparison and generate a largescale ranking set to pre-train the network; For "annotation sticker", the absolute noise-containing subjective scores are transformed into ranking comparison results, and we design an indirect unsupervised regression based on Eigenvalue Decomposition (EVD); For "distortion sticker", we propose a perception-based distortion classification method, which makes the distortion types clear and refined. Experiments have proved that our NR IQA approach Experiments show that the algorithm performs well and has good generalization ability. Furthermore, the proposed perception based distortion classification method would be able to provide insights on how the visual related studies may be developed and to broaden our understanding of human visual system.

A Framework for False Image Detection with Sample-Oriented Intelligent Adversarial

To develop algorithms capable of automatically detecting and evaluating the authenticity of images and videos, researchers have focused on false image detection algorithms. These algorithms aim to identify the authenticity of images by distinguishing between real images and forged ones generated using false generation algorithms. In this paper, the main focus is on implementing a single-frame authenticated image detector using the concept of migration learning. The detector utilizes Inception ResNet v2, a target classification network pre-trained on a self-built military scene dataset. To enhance the dataset, a series of graphical enhancement algorithms are employed, enabling the classification network to learn the crucial differences between real and forged images. Additionally, Focal Loss is introduced to balance the dataset for various GAN-based image generation algorithms. As a result, the final forged target image detector achieves an impressive classification accuracy of 0.8908 on a large-scale sample test set.

An Adaptive Dimension Weighting Spherical Evolution to Solve Continuous Optimization Problems

The spherical evolution algorithm (SE) is a unique algorithm proposed in recent years and widely applied to new energy optimization problems with notable achievements. However, the existing improvements based on SE are deemed insufficient due to the challenges arising from the multiple choices of operators and the utilization of a spherical search method. In this paper, we introduce an enhancement method that incorporates weights in individuals’ dimensions that are affected by individual fitness during the iteration process, aiming to improve SE by adaptively balancing the tradeoff between exploitation and exploration during convergence. This is achieved by reducing the randomness of dimension selection and enhancing the retention of historical information in the iterative process of the algorithm. This new SE improvement algorithm is named DWSE. To evaluate the effectiveness of DWSE, in this study, we apply it to the CEC2017 standard test set, the CEC2013 large-scale global optimization test set, and 22 real-world problems from CEC2011. The experimental results substantiate the effectiveness of DWSE in achieving improvement.

A large-scale last-mile consolidation model for e-commerce home delivery

E-commerce's rapid growth, combined with customer demand for fast shipping, has significantly escalated last-mile transportation, particularly home deliveries. This surge calls for novel strategies to optimize vehicle utilization while maintaining timely delivery. In response, many online retailers are incentivizing customers to embrace delayed home deliveries by offering economic incentives. Nonetheless, current transportation systems fall short in accommodating orders with extended delivery windows, primarily adhering to the First-In-First-Out rule and distance minimization via the Vehicle Routing Problem (VRP). In this paper, we introduce a new consolidation-based delivery methodology that addresses these challenges. Our approach accounts for orders with daily-based time windows and also anticipates future demand realizations (i.e., expected incoming orders). We characterize this consolidation problem using a Mixed Integer Linear Programming model and propose a custom metaheuristic approach that can tackle the problem on a large-scale setting, introducing a significant novelty approach in last-mile delivery research. We applied our methodology to one of Mexico's largest retailers and compared its performance against the company's existing transportation systems. Our approach substantially improves vehicle utilization and yields considerable reductions in distance traveled, time, and overall transportation costs, achieving cost savings of up to 52%. These savings represent tangible benefits, enabling potential revenue enhancements for businesses and cost-effective, timely deliveries for consumers. Additionally, the increased vehicle utilization implies fewer vehicles are needed for the same volume of deliveries, thereby enhancing operational efficiency. This innovative approach, therefore, presents a practical and highly effective solution for managing large-scale last-mile delivery scenarios.

Accelerated Design of Ultra-High-Performance Aramid Copolymers via a High-Throughput Screening Approach.

Developing advanced materials, such as functional polymers, poses a significant challenge as a result of the vastness of the material space that needs to be explored, which could potentially be infinite in principle. We propose a data-driven high-throughput screening approach coupled with molecular dynamics (MD) simulations to address this issue in the design of high-performance co-polymerized aramid fibers. We aimed to identify diamine monomers that could replace 3,4'-oxydianiline in Technora from a large-scale set (1 920 304) of possible monomers that were prepared from the PubChem database. We initially screened these monomers using a cheminformatics-based approach, considering four criteria: complexity, neutrality, linearity, and gyration radius of the molecule. Then, we performed subsequent screening based on MD simulations to estimate interchain interaction energies under both stretched and melted conditions and tensile strength simulations. Our screening approach successfully identified 31 promising and novel diamine monomers for aramid copolymers. This demonstrates the potential and effectiveness of our approach as a promising protocol for exploring targeted chemical spaces in designing novel monomers for high-performance aramid fibers and possibly other advanced polymers.