Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,341 papers · 148 categories

Trend · papers per month

146293439585 · Jun 202019922001200920182026
48 results for internet applications

Paper uses t-SNE to classify China's Internet finance risks.

problem Systemic risk characteristics of China's Internet finance during macroeconomic shocks.
method t-SNE machine learning algorithm for data mining and risk classification.
result Identified peak and thick-tail characteristics of Internet financial systemic risk.

Study uses trillion internet observations to analyze social science insights.

problem Understanding social science insights from internet data.
method Unified dataset of over 1.5 trillion observations, applied to urban growth, sleep duration, and economic productivity.
result Internet growth reaches saturation at 1 IP per 3 people, taking 16.1 years.

This work tackles multi-objective optimization using constrained formulations for internet applications.

problem Handling user and item level constraints, scalability, and variance of dual estimates.
method Constrained optimization formulations, Lagrangian duality, user and item level constraints handling, scalability improvements, variance reduction.
result Significant advancement in state-of-the-art multi-objective optimization for internet applications.

The study finds that Chinese internet users have different search behaviors and attention patterns.

problem Heterogeneity in search behavior and attention among Chinese internet users.
method Data extraction technology to analyze Baidu Index keyword search volume data.
result Chinese internet users exhibit different search behaviors and attention patterns.

Nemesyst optimizes deep learning for IoT food refrigeration systems.

problem Optimizing deep learning for large-scale IoT systems with dynamic data.
method Hybrid parallelism framework using databases and model sequentialisation.
result Demonstrated compelling performance of deep learning models in IoT food refrigeration.

Locality-sensitive hashing speeds up web app security testing.

problem Challenges in crawling Rich Internet Applications (RIAs) due to state similarity.
method Uses MinHash sketches to analyze DOM structures and estimate similarity.
result Successfully scans RIAs that traditional crawling methods cannot.

Study predicts internet-based treatment effects for GPPPD based on dyadic coping.

problem Identifying which patients will benefit most from internet-based GPPPD treatment.
method Developed a multivariable decision tree model using recursive partitioning.
result Predicts large effects for high dyadic coping patients, small effects for low dyadic coping patients.

This paper is the first attempt to formalize a new field of economics; studding the Intangibles Goods available on the Internet. We are taking advantage of the digital world's specific rules, in particular the zero marginal cost, to propose a theory of trading & sharing unified. A function based money is created as a w…

2014-05-07abs ↗pdf ↗

Automated system forecasts safety and security issues in internet-connected devices.

problem Challenges in real-time large-scale forecasting by human experts.
method Developed a scalable and versatile forecasting system using models and methods.
result Automated short and long-term forecasts detect safety and security issues in internet-connected devices.

This article reviews ML applications in wireless networks over 30 years.

problem Achieving high-rate, low-latency, and reliable wireless services in complex networks.
method Exploration of various ML algorithms including supervised, unsupervised, reinforcement, and deep learning.
result Machine learning algorithms have been successfully applied to wireless networks.

Fault detection in IoT nodes using Dempster-Shafer Theory of Evidence.

problem Fault detection in sensor nodes in IoT applications.
method Dempster-Shafer Theory of Evidence for sensor data fusion.
result Competent accuracy for fault detection in both benchmark and laboratory data sets.

Extracts patterns from mobile network data for better resource management.

problem Improving network efficiency and resource allocation for mobile users.
method Spatiotemporal analysis of internet activity records (IARs) data.
result Developed a mobile traffic partitioning scheme.

Paper proposes a deep autoencoder model to detect anomalies in CAV locations.

problem Early detection of anomalies in self-reported vehicle locations for CAVs.
method Unsupervised learning model based on deep autoencoder using vehicle locations and RSSI.
result The proposed model is effective and robust in detecting self-reported location anomalies.

FinGPT democratizes financial data for LLMs, enabling innovation.

problem Limited financial text datasets and disparities between general and financial text data.
method Automates collection and curation of real-time financial data from diverse Internet sources, fine-tuning with RLSP and LoRA.
result Democratizes access to financial data for LLMs, enabling innovation.

Mobile sensing is an emerging technology that utilizes agent-participatory data for decision making or state estimation, including multimedia applications. This article investigates the structure of mobile sensing schemes and introduces crowdsourcing methods for mobile sensing. Inspired by social network, one can estab…

2015-08-03abs ↗pdf ↗

Researchers develop a machine learning model to improve smell perception for food quality monitoring.

problem Limited data in target domain makes it hard to accurately classify food quality.
method Weakly supervised domain adaptation framework using multiple models in supervised and unsupervised settings.
result The approach generalizes well from one domain to another, improving food quality classification.

Deep learning model detects and corrects outliers in crowd-sourced weather data.

problem Data quality issues in crowd-sourced weather data.
method Bayesian deep learning approach with Gaussian-uniform mixture density network.
result Automated outlier detection in spatio-temporal environmental modeling.

A novel approach reduces class imbalance in network traffic classification.

problem Severe class imbalance in network traffic leads to poor classification performance.
method Group & Reweight strategy: clusters classes, updates weights, optimizes model.
result Improves comprehensive performance in prediction and reduces class imbalance.

In this paper we present a review of the existing typologies of Internet service users. We zoom in on social networking services including blogs and crowdsourcing websites. Based on the results of the analysis of the considered typologies obtained by means of FCA we developed a new user typology of a certain class of I…

2013-11-30abs ↗pdf ↗

Privacy-preserving offloading for deep learning on IoT devices.

problem Resource constraints and privacy risks in deep neural network computations on IoT devices.
method Obfuscation of data before offloading to a remote node, allowing accurate classification by trained neural networks.
result Data obfuscation enables accurate neural network classification without exposing sensitive information.

Feedback loops amplify dataset biases, affecting future model performance.

problem Feedback loops amplify biases in datasets, risking future model reliability.
method Formalized system where model interactions are recorded and reused, analyzed for bias amplification.
result Models that behave like samples from the training distribution are more stable and calibrated.

Paper develops security model and pricing for stable digital currency in quantum blockchain network.

problem Securing and pricing stable digital currency in a quantum blockchain network.
method Developed a block-based quantum channel networking technology and a FinTech platform model with dynamic pricing.
result Established a generalized IoB security model using quantum channel networking and QKD.

IoT nodes compress measurements into DNN outputs for efficient communication.

problem Efficient communication of high-dimensional IoT data with limited bandwidth.
method Modeling IoT node measurements as DNN intermediate outputs and optimizing model parameters.
result Approximately 96% reduction in transmissions with only 2.5% loss in inference accuracy.

Geometric approach monitors dynamic large graphs, detects major events.

problem Monitoring dynamic large graphs is challenging due to local changes affecting global properties.
method Developed a geometric approach using Ollivier-Ricci curvature for real-time monitoring.
result Detects major events and changes via graph embedding geometry.

Study shows best hyper-parameters improve deep learning model's accuracy for IoT attack detection.

problem Improving accuracy of deep learning model for IoT attack detection.
method Examined three hyper-parameters' influence on model performance.
result Model's reported accuracy not achievable due to optimal hyper-parameters.

Machine learning predicts COVID-19 activity in China.

problem Real-time forecasting of COVID-19 activity in Chinese provinces.
method Combines mechanistic disease models with digital traces (internet searches, news alerts). Uses clustering and data augmentation techniques.
result Stable and accurate forecasts 2 days ahead of current time, outperforming baseline models in 27 out of 32 provinces.

Bayesian model identifies outliers and determines tensor rank in streaming data.

problem Outliers and over-fitting in streaming tensor factorization.
method Variational Bayesian Inference for robust tensor rank determination and outlier identification.
result Model accurately identifies sparse outliers and determines tensor rank.