Teleoscope enables thematic curation of large Reddit corpora via interactive refinement, with three deployments indicating benefits in serendipitous keyword discovery, search saturation confidence, and collaborative curation discussions.
Title resolution pending
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
representative citing papers
Introduces a parallel joint symbolic approximation technique for large-scale time series via local-global decoupling that maintains reconstruction quality with reduced runtime.
A CBR system based on similarity of local explanations provides visualizations that fraud analysts at a Dutch bank found useful and easy to use for processing ML-generated fraud alerts.
Human-grounded evaluation finds no significant performance improvement from adding SHAP explanations to model confidence scores in alert processing.
A seq2seq model is proposed to learn universal embeddings from wearable and ambient sensor data for ADL recognition and semi-supervised learning.
A methodology is proposed to detect class-based concept drift by self-evaluating predictive model degradation, with experiments on synthetic and real-world datasets showing effectiveness.
citing papers explorer
-
Crystallizing Schemas with Teleoscope: Thematic Curation of Large Text Corpora on Reddit
Teleoscope enables thematic curation of large Reddit corpora via interactive refinement, with three deployments indicating benefits in serendipitous keyword discovery, search saturation confidence, and collaborative curation discussions.
-
Parallel Two-Stage Approach for Joint Symbolic Approximation of Time Series
Introduces a parallel joint symbolic approximation technique for large-scale time series via local-global decoupling that maintains reconstruction quality with reduced runtime.
-
Case-Based Reasoning for Assisting Domain Experts in Processing Fraud Alerts of Black-Box Machine Learning Models
A CBR system based on similarity of local explanations provides visualizations that fraud analysts at a Dutch bank found useful and easy to use for processing ML-generated fraud alerts.
-
A Human-Grounded Evaluation of SHAP for Alert Processing
Human-grounded evaluation finds no significant performance improvement from adding SHAP explanations to model confidence scores in alert processing.
-
Activity2Vec: Learning ADL Embeddings from Sensor Data with a Sequence-to-Sequence Model
A seq2seq model is proposed to learn universal embeddings from wearable and ambient sensor data for ADL recognition and semi-supervised learning.
-
Automating concept-drift detection by self-evaluating predictive model degradation
A methodology is proposed to detect class-based concept drift by self-evaluating predictive model degradation, with experiments on synthetic and real-world datasets showing effectiveness.
- Deep Learning using Rectified Linear Units (ReLU)