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.

169,181 papers · 148 categories

Trend · papers per month

count % of corpus

12 results for “ecommerce”

Detecting organized eCommerce fraud using scalable categorical clusteringstat.ML

New method detects organized eCommerce fraud by clustering orders.

problem Detecting coordinated fraud by groups of fraudsters.

method Scalable categorical clustering using agglomerative clustering and sampling.

result Groups 35-45% of fraudulent orders together, detects 26.2% of fraud with low false alarms.

fraud detection agglomerative clustering sampling organized fraud eCommerce

2019-10-10abs ↗pdf ↗

Large Scale Product Categorization using Structured and Unstructured Attributescs.IR

Paper tackles product categorization with structured and unstructured attributes for large-scale eCommerce.

problem Challenges in categorizing products with thousands of classes and millions of products.

method Compares hierarchical and flat models, uses Deep Learning for feature extraction, combines structured and unstructured attributes.

result Flat models perform better in specific cases, and the proposed approach handles faulty attribute names and values.

product categorization structured attributes unstructured attributes Deep Learning convolutional filters

2019-03-01abs ↗pdf ↗

Predicting purchasing intent: Automatic Feature Learning using Recurrent Neural Networkscs.LG

Neural network predicts purchasing intent without feature engineering.

problem Predicting purchasing intent in ecommerce with minimal feature engineering.

method Trainable vector spaces, multi-layer recurrent neural networks, parameter sharing, skip connections.

result Classification accuracy exceeds state-of-the-art on benchmark datasets.

neural networks purchasing intent ecommerce feature learning recurrent neural networks

2018-07-21abs ↗pdf ↗

A Machine Learning Approach to Shipping Box Designstat.ML

Machine learning optimizes box selection for efficient warehouse operations.

problem Optimizing the selection of shipping boxes for thousands of products.

method Formulated as a clustering problem, parameters estimated via analytics.

result Improves box utilization rate by over 10%.

machine learning shipping box design fulfillment clustering

2018-09-26abs ↗pdf ↗

An Empirical Comparison of FAISS and FENSHSES for Nearest Neighbor Search in Hamming Spacecs.LG

This paper compares FAISS and FENSHSES for nearest neighbor search in Hamming space.

problem Comparing nearest neighbor search systems in Hamming space.

method Comprehensive evaluations of indexing speed, search latency, and RAM consumption.

result Better understanding of trade-offs between main memory and secondary memory systems.

nearest neighbor search Hamming space FAISS FENSHSES trade-offs

2019-06-24abs ↗pdf ↗

How to Grow a (Product) Tree: Personalized Category Suggestions for eCommerce Type-Aheadcs.LG

SessionPath improves category suggestions in type-ahead search.

problem Improving precision and recall in eCommerce type-ahead suggestions.

method SessionPath uses session embeddings and a probability distribution model to predict facets.

result SessionPath outperforms count-based and neural models in eCommerce shops.

E-commerce Type-ahead search Facet suggestions Session embeddings Probability distribution

2020-05-26abs ↗pdf ↗

A compendium of data sources for data science, machine learning, and artificial intelligencecs.LG

Provides a compendium of data sources for various applications.

problem Lack of comprehensive data sources for data science, machine learning, and AI.

method Compilation of diverse data sources across multiple application areas.

result A comprehensive list of data sources for data scientists and machine learning experts.

data sources data science machine learning AI compendium

2023-09-10abs ↗pdf ↗

Multi-Armed Bandit Strategies for Non-Stationary Reward Distributions and Delayed Feedback Processescs.LG

Survey of MAB strategies for non-stationary reward distributions with delayed feedback.

problem Optimizing product availability in an online grocery pick-up platform with non-stationary and delayed reward feedback.

method Evaluation of

ε

-greedy, UCB1, Thompson Sampling, and a new adaptive technique (AG1) in MAB simulations.

result AG1 outperforms traditional MAB strategies in minimizing regret for non-stationary and delayed feedback.

Multi-Armed Bandit Non-Stationary Reward Delayed Feedback Regret Minimization E-commerce

2019-02-22abs ↗pdf ↗

Online Evaluation of Audiences for Targeted Advertising via Bandit Experimentscs.LG

Adaptive algorithm for online evaluation of targeted audiences in advertising.

problem Determining the right match between advertising creatives and target audiences.

method Contextual bandit approach to address audience overlap and learn optimal display policies.

result The proposed method is more efficient than traditional split-testing methods.

online evaluation targeted advertising bandit algorithms audience overlap creative matching

2019-07-04abs ↗pdf ↗

Fast Botnet Detection From Streaming Logs Using Online Lanczos Methodcs.LG

Paper introduces fast botnet detection from web server logs.

problem Botnets cause various malicious activities; fast detection is needed.

method Inspired by PCA, uses Lanczos method to improve detection speed.

result Lanczos method significantly reduces time complexity for botnet detection.

botnet detection streaming logs PCA Lanczos method web server logs

2018-12-19abs ↗pdf ↗

Towards automated feature engineering for credit card fraud detection using multi-perspective HMMscs.LG

A new approach models credit card transactions using HMMs to detect fraud.

problem Detecting credit card fraud using isolated event analysis.

method Model sequences from three perspectives using HMMs and combine likelihoods as features.

result Improved fraud detection effectiveness compared to state-of-the-art methods.

credit card fraud HMMs feature engineering sequence analysis random forest

2019-09-03abs ↗pdf ↗

Causally Driven Incremental Multi Touch Attribution Using a Recurrent Neural Networkcs.LG

A system for attributing ad effects using a neural network and Shapley values.

problem Attributing ad effects to individual ads in a complex, sequential environment.

method A two-step approach: response modeling with RNN and credit allocation with Shapley values.

result The system accurately allocates incremental ad effects to individual ads, handling sequence dependence.

Multi Touch Attribution Recurrent Neural Network Shapley Values Incremental Attribution Sequential Data

2019-02-01abs ↗pdf ↗