FANDA: A Novel Approach to Perform Follow-Up Query Analysis
Authors: Qian Liu, Bei Chen, Jian-Guang Lou, Ge Jin, Dongmei Zhang6770-6777
AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | The experimental results on Follow Up demonstrate the superiority of FANDA over multiple baselines across multiple metrics. We conduct experimental studies on the Follow Up dataset. Multiple baselines and metrics are utilized to demonstrate promising results of our model. |
| Researcher Affiliation | Collaboration | Qian Liu, Bei Chen, Jian-Guang Lou, Ge Jin, Dongmei Zhang Beihang University, Beijing, China Microsoft Research, Beijing, China Peking University, Beijing, China qian.liu@buaa.edu.cn; {beichen, jlou, dongmeiz}@microsoft.com; elvisking@pku.edu.cn |
| Pseudocode | No | The paper describes the FANDA approach conceptually and with flow diagrams (Figure 1), but does not include formal pseudocode or an algorithm block. |
| Open Source Code | No | The paper states the 'Follow Up' dataset is 'Available at https://github.com/SivilTaram/FollowUp', but does not explicitly state that the source code for the FANDA methodology itself is available there or elsewhere. |
| Open Datasets | Yes | We build a new dataset named Follow Up 1,which contains 1000 query triples on 120 tables. To the best of our knowledge, it is the first public dataset that contains various kinds of follow-up scenarios. Available at https://github.com/SivilTaram/FollowUp |
| Dataset Splits | Yes | We evaluate our methods on the proposed Follow Up dataset, and split the 1000 triples following the sizes 640/160/200 in train/development/test. |
| Hardware Specification | No | The paper states 'We trained our models on a GPU' but does not provide specific details such as GPU model, CPU type, or memory specifications. |
| Software Dependencies | No | The paper mentions using 'Spacy' and 'Glove' but does not provide specific version numbers for these or any other software dependencies. |
| Experiment Setup | Yes | In query utterances, the words can be divided into two types: analysis-specific words and rhetorical words... As shown in Table 2, we predefine 8 types of symbol for different analysis-specific words. Given a query, anonymization is to recognize all analysis-specific words in it, and replace them with the corresponding symbols to construct a symbol sequence. ...The compositional deduction rules originate from SQL clause syntax... φ is an embedding function initialized using Glove (Pennington, Socher, and Manning 2014). ...where > 0 is a hyperparameter. |