DA-bench

Visual Benchmark for Data Analytics AI Agents

juliusRun #150 (December 6, 2024)

Overall Score: 53.5%
(40% Setup Score + 60% Test Score)

DA-bench Setup for Run #150 — Setup Score: 50.0 (3 / 6)

See how this tool was set up for the test.

  • Unchecked Connects to Data Warehouse
  • Verified No Individual Upload of Files
  • Verified No SQL for Setup
  • Verified Setup Less Than Ten Minutes

DA-bench Results for Run #150 — Test Score: 55.8 (120 / 215)

Data Querying (85 / 120)
Question Date Tested Overall Score Video Recording

dq01
Perform an aggregation on an explicit column

December 6,2024 5

dq02
Perform an aggregation with an explicit table but not an inferred column

December 6,2024 5

dq03
Perform an aggregation with an implicit table and implicit column

December 6,2024 5

dq04
Find and compare information across tables without joins

December 6,2024 5

dq05
Work with non-literal values

December 6,2024 5

dq06
Work with non-literal values and non-SQL data manipulation

December 6,2024 5

dq07
Deal with common acronymns and more advanced aggregations

December 6,2024 5

dq08
Multi-step queries

December 6,2024 -5

dq09
Aggregations with numeric predicates to filter

December 6,2024 5

dq10
Aggregations with categorical predicates to filter

December 6,2024 5

dq11
Schema review

December 11,2024 5

dq12
Aggregate records that are filtered with a predicate requiring a join

December 6,2024 5

dq13
Recognizes truly ambiguous queries.

December 6,2024 -5

dq14
Can handle boolean features

December 6,2024 5

dq15
Handles ambiguous column names

December 6,2024 5

dq16
Understands set operates require consideration of overlap

December 6,2024 5

dq17
Finds relevant values inside a Column to answer questions

December 6,2024 5

dq18
Lookup a single record by ID

December 6,2024 5

dq19
Perform an aggregation by a different name and a second query from that

December 6,2024 5

dq20
Perform an aggregation based on a very different question name

December 6,2024 5

dq21
Perform a filter and an unusually-phrased aggregation in the correct order

December 6,2024 -5

dq23
Can handle incorrect column names well

December 6,2024 5

dq24
Schema review

December 6,2024 0

dq25
Work with non-literal values

December 6,2024 5
Domain Knowledge (5 / 5)
Question Date Tested Overall Score Video Recording

dk01
Column Relevance Determination

December 6,2024 5
Feature Engineering (20 / 40)
Question Date Tested Overall Score Video Recording

fe1
Make a boolean indicator feature for a criteria set

December 6,2024 5

fe2
Make a categorical feature from a criteria set

December 6,2024 5

fe3
Minmax normalization

December 6,2024 5

fe4
Combining two input columns

December 6,2024 5

fe5
Sentiment

December 6,2024 -5

fe6
Phrase Identification in Text

December 6,2024 -5

fe7
Advanced NLQ

December 6,2024 5

fe8
Advanced NLQ

December 6,2024 5
Insight Identification (15 / 25)
Question Date Tested Overall Score Video Recording

ii2
Compare an aggregation for two distinct subsets of data

December 6,2024 5

ii5
Identifying basic trends on short timelines

December 6,2024 5

ii6
Understands statistical significance

December 6,2024 5

ii7
Understands derivitives

December 6,2024 -5

ii8
Can use NLQ feature engineering as part of an insight request

December 6,2024 5
Learning (10 / 10)
Question Date Tested Overall Score Video Recording

l1
Can remember the meanings of oddly-named columns

December 6,2024 5

l2
Can remember criteria sets under a single name

December 6,2024 5
Visualization (-15 / 15)
Question Date Tested Overall Score Video Recording

v1
Basic Charting

December 6,2024 -5

v2
Charting with two series

December 6,2024 -5

v3
Categorical Charts

December 6,2024 -5