Intent Detection
Intent Detection is a task of determining the underlying purpose or goal behind a user's search query given a context. The task plays a significant role in search and recommendations. A traditional approach for intent detection implies using an intent detector model to classify user search query into predefined intent categories, given a context. One of the key challenges of the task implies identifying user intents for cold-start sessions, i.e., search sessions initiated by a non-logged-in or unrecognized user.
Source: Analyzing and Predicting Purchase Intent in E-commerce: Anonymous vs. Identified Customers
Papers
Showing 1–10 of 330 papers
All datasetsATISMixSNIPSMixATISSNIPSASOS.com user intentBANKING77ATIS (vi)BANKING77 10-shotBANKING77 5-shotCAISCLINC150CLINC150 10-shot
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Bi-model with decoder | Accuracy | 98.99 | — | Unverified |
| 2 | Transformer-Capsule | Accuracy | 98.89 | — | Unverified |
| 3 | Attention Encoder-Decoder NN | Accuracy | 98.43 | — | Unverified |
| 4 | Joint model with recurrent slot label context | Accuracy | 98.4 | — | Unverified |
| 5 | CTRAN | Accuracy | 98.07 | — | Unverified |
| 6 | Joint BERT + CRF | Accuracy | 97.9 | — | Unverified |
| 7 | SF-ID | Accuracy | 97.76 | — | Unverified |
| 8 | SF-ID (BLSTM) network | Accuracy | 97.76 | — | Unverified |
| 9 | Joint BERT | Accuracy | 97.5 | — | Unverified |
| 10 | JointBERT-CAE | Accuracy | 97.5 | — | Unverified |