Dataset Viewer
Auto-converted to Parquet Duplicate
text
stringlengths
2
82
intent
class label
17 classes
script
stringclasses
4 values
প্রধানমন্ত্রী কে .
16out_of_scope
bn
okay ashi tahole
1goodbye
bl
একটু বাংলাদেশের রাজধানী কি
16out_of_scope
bn
একটু রাখছি এখন
1goodbye
bn
service er man din din kharap hocche?
14complaint
bl
hi how are you
0greeting
en
সমস্যা সমাধান হয়ে গেছে ধন্যবাদ
2thanks
bn
minimum balnce koto ache amar
3balance_inquiry
bl
gato mser statement din
4transaction_history
bl
apnader biruddhe bangladesh bank e obhijog করবো
14complaint
mx
current balance কত
3balance_inquiry
mx
একটু রংপুর ব্রাঞ্চের ফোন নম্বর দিন
12branch_atm_info
bn
আচ্ছা আমার অ্যাকাউন্ট থেকে টাকা তোলা যাচ্ছে না
10account_issue
bn
new card
7card_issue
en
ki ki lagbe
9account_opening
bl
Manusher support dorkar ekhoni
15agent_request
bl
eto sundor করে bujhanor jonno dhonnobad
2thanks
mx
হয়রানি একটু
14complaint
bn
কথা বলতে পারি??
0greeting
bn
অনলাইন ট্রান্সফার কি ফ্রি
13charges_fees
bn
app evul password dekhacche barbar
11login_issue
bl
onek help korlen
2thanks
bl
nominee poriborton krte chai
10account_issue
bl
how do i reset my password
11login_issue
en
obhijog
14complaint
bl
accha kyc update korte hobe ektu
10account_issue
bl
Call me
15agent_request
en
my 2000 taka isstuck
6transaction_failed
en
ট্রানজেকশন পেন্ডিং দেখাচ্ছে অনেকক্ষণ
6transaction_failed
bn
টাকা atke geche ki korbo
6transaction_failed
mx
card unblock করুন
7card_issue
mx
কত বছরের জন্য লোন দেন
8loan_inquiry
bn
kivabe biriyani ranna kore
16out_of_scope
bl
where isyour head office
12branch_atm_info
en
সেভিংস অ্যাকাউন্ট এ ব্যালেন্স কত দয়া করে
3balance_inquiry
bn
অ্যাকাউন্ট বন্ধ ভাই
10account_issue
bn
আচ্ছা বিকাশ এ টাকা পাঠাবো কিভাবে
5fund_transfer
bn
Bot na ami manush chai
15agent_request
bl
dhonnobad apnake onek
2thanks
bl
app e vul password dekhacche barbar
11login_issue
bl
একটু স্যালারি অ্যাকাউন্ট খুলবো কিভাবে আপু
9account_opening
bn
hellu is anyone online
0greeting
en
হ্যালো ভাই কথা বলা যাবে
0greeting
bn
500 taka onno bank e pthan
5fund_transfer
bl
what is a securd loan
8loan_inquiry
en
Kivabe biriyani ranna kore
16out_of_scope
bl
মানুষের সাপোর্ট দরকার এখনই .
15agent_request
bn
ektu card block vai
7card_issue
bl
আমার 1500 টাকা আটকে আছে প্লিজ
6transaction_failed
bn
is the bank open today
12branch_atm_info
en
card unblck korun
7card_issue
bl
ট্রান্সফার ফেইল হয়েছে টাকা কাটা হয়েছে
6transaction_failed
bn
bye bye
1goodbye
bl
excuse me i will close my account this service is so bad sir
14complaint
en
kal abarashbo
1goodbye
bl
nagad e taka pathabo kivabe
5fund_transfer
bl
আচ্ছা একদিনে সর্বোচ্চ কত পাঠানো যায়
5fund_transfer
bn
thats all for now
1goodbye
en
কিস্তি .
8loan_inquiry
bn
মিরপুর এ আপনাদের ব্রাঞ্চ কোথায়??
12branch_atm_info
bn
শুনেন কোথায় কোথায় টাকা খরচ হয়েছে দেখান আপু
4transaction_history
bn
বরিশাল তে এটিএম আছে কি??
12branch_atm_info
bn
Sylhet branch er phone number din
12branch_atm_info
bl
taka katar sms esechekintu balance thik nai
6transaction_failed
bl
shunen app cholche na apu
11login_issue
bl
gato 2 diner lnden dekhan
4transaction_history
bl
I cannot withdraw from my account
10account_issue
en
শুনেন আসসালামু আলাইকুম ভাই
0greeting
bn
how much is the sms alert charge
13charges_fees
en
my nameis wrong on the account
10account_issue
en
is there a charge for early loan repayment
8loan_inquiry
en
minimum balance kotoache amar
3balance_inquiry
bl
amar account er balance dekhan??
3balance_inquiry
bl
আমি আপনাদের ব্যাংকে নতুন তাই একটু সাহায্য করুন, কে আমাকে টাকা পাঠিয়েছে দেখতে চাই
4transaction_history
bn
hello vai amar account freeze kora hoyeche keno
10account_issue
bl
thanks so much for helping asap
2thanks
en
can you email me the statement
4transaction_history
en
খোলা আছে আপু
12branch_atm_info
bn
hello nice to meet you
0greeting
en
কাস্টমার কেয়ার .
15agent_request
bn
hey is it open asap
12branch_atm_info
en
ভ্যাট কত কাটা হয় লেনদেনে রিপ্লাই দিবেন প্লিজ
13charges_fees
bn
so harassment please
14complaint
en
realperson please
15agent_request
en
Dhonnobad
2thanks
bl
virtual card pabo কিভাবে
7card_issue
mx
shunen apnader biruddhe bangladesh bank e obhijog korbo
14complaint
bl
amar login lock hoie geche
11login_issue
bl
শুনেন এটিএম কার্ড আটকে গেছে মেশিনে ভাই
7card_issue
bn
ভাই কার্ডে অচেনা লেনদেন দেখাচ্ছে ব্লক করুন প্লিজ
7card_issue
bn
i want to do a fund trnsfer
5fund_transfer
en
I want to open a dps
9account_opening
en
আচ্ছা নতুন কার্ড কিভাবে পাবো
7card_issue
bn
Balance inquiry korte chai
3balance_inquiry
bl
can you email me the statement thank you
4transaction_history
en
খারাপ ব্যবহার
14complaint
bn
cheque diye taka joma dibo kivabe
5fund_transfer
bl
fail
6transaction_failed
bl
i want my money back for the failed transaction
6transaction_failed
en
myatm card is not working
7card_issue
en
End of preview. Expand in Data Studio

Bangla / English / Banglish Banking Intent Classification

A 17-intent classification dataset for a Bangladeshi retail-banking chatbot, covering the three ways customers actually write:

script example rows
bn Bengali script আমার ব্যালেন্স কত 1,817
en English what is my balance 1,817
bl Banglish (romanized Bangla) amar balance koto 1,700
mx code-mixed mid-sentence taka katlo kintu transfer hoyni নাই 353

5,687 rows, 17 intents — including an explicit out_of_scope reject class.

⚠️ This is synthetic data. It is a bootstrap for getting a CPU intent classifier off the ground when you have no logs yet, not a substitute for real ones. See Limitations before you rely on a number measured here. The companion hand-written holdout is the honest signal.

Dataset structure

Fields

field type description
text string the user message, 1–16 words
intent class_label one of 17 labels (below)
script string bn | en | bl | mx — writing system, useful for per-script error analysis

script is metadata, not a training feature. It exists so you can report accuracy per writing system, which is where the interesting failures hide — Banglish and code-mixed rows are consistently harder than either monolingual form.

Splits

from datasets import load_dataset
ds = load_dataset("Badhon/BanglaBankingIntent")
# DatasetDict({train: 4277, validation: 692, test: 718})

The splits are disjoint at template level, not row level. Each template is assigned to exactly one split before it expands into surface rows, so no test row is a respelling, recasing, code-mixing or politeness-affixed variant of a training row. Leakage is also blocked on a punctuation/case/affix-insensitive canonical form, so balance in train does not permit Balance?? in test.

A dataset built the naive way — expand first, split rows randomly — reports ~99.9% test accuracy that is pure memorization. Under template-level splitting the shipped transformer scores 0.715 on test and 0.518 on the hand-written holdout. That gap is the honest measure of how much of the test score is convention-following rather than generalization.

Label distribution

intent train val test total description
fund_transfer 325 44 49 418 wants to move money now — send/transfer/pay
balance_inquiry 305 37 44 386 a read of current state — "how much is in there"
card_issue 291 48 46 385 card blocked, lost, stolen, PIN, activation, expiry
loan_inquiry 295 43 47 385 loans, EMI, interest rates, eligibility, repayment
branch_atm_info 268 60 49 377 where is a branch/ATM, hours, is it open
transaction_failed 280 38 43 361 a specific payment debited but did not arrive, or was declined
greeting 261 49 47 357 opener, whole message
goodbye 264 46 46 356 sign-off
account_opening 249 42 53 344 wants to open a new account; documents, minimum deposit
transaction_history 268 35 33 336 a read of past events — statements, "last 5 transactions"
thanks 230 32 36 298 gratitude, whole message
complaint 221 37 39 297 grievance with no specific remedy asked
charges_fees 217 36 41 294 maintenance fee, transfer charge, annual fee, excise duty
account_issue 207 44 39 290 an existing account is frozen, dormant, locked, KYC expired
login_issue 211 38 41 290 cannot get into the app — password, OTP, app PIN reset
out_of_scope 206 34 38 278 chitchat, other domains, noise
agent_request 179 29 27 235 escalate to a human

Roughly balanced by design (per-intent row caps during generation).

Label boundaries

Several intents share vocabulary (taka, account, transaction) and differ only in what the user wants done. The tie-breaks used to label consistently, documented in full in the domains/banking.py docstring:

  • balance_inquiry vs transaction_history — current state vs past events. A question about one specific transaction that went wrong is transaction_failed, not history.
  • fund_transfer vs transaction_failed — wants to move money now vs the money already moved and is missing. The defining feature of transaction_failed is a broken transaction, not a general grievance.
  • login_issue vs card_issue — app/internet-banking access (password, OTP, app PIN) vs the physical/virtual card (including card PIN). This split is deliberate and is the most common labeling mistake in this set.
  • account_issue vs account_opening — an existing account is broken vs wants a new one.
  • complaint — angry with no actionable request that fits above. If the user is angry and names a failed transfer, label transaction_failed: the actionable intent wins.

Rule that overrides all of the above: a greeting glued onto a real request is labeled by the request, never the greeting. assalamu alaikum vai amar card block hoye gese is card_issue.

out_of_scope

The reject class, and the reason to prefer this dataset over a 16-intent one. A closed-set softmax must put ~1.0 of its probability mass on some label, so a model without a reject class answers tomar basa kothay? as a confident complaint. No confidence threshold fixes that, because the model was never given a way to express "none of the above".

Coverage spans bot-directed chitchat (tumi ki manush), other industries and domains (weather, cricket, prayer times, politics), general-assistant requests (write a poem, do this maths), and meta/noise (test test, keyboard mash, emoji-only, hmm).

Deliberately not out_of_scope: profanity aimed at the bank (that is complaint — actionable, route to a human), and vague-but-financial fragments (koto ache? is balance_inquiry).

The class is capped at the same size as the others on purpose. An oversized reject class raises the false-fallback rate — real customers routed to "I don't understand" — which costs more in production than a missed rejection.

Evaluation

Do not report the test split alone. It is template-disjoint from train, which makes it honest, but it still only answers "can you generalize across our own templates". Pair it with the hand-written banking holdout (156 items, not shipped as a split because it must never be trained on):

INTENT_DOMAIN=banking python transformer_model/eval_holdout.py

Every holdout item is written by hand to share no template with the generated data, and the generator enforces this: any generated row matching a holdout item is dropped at source, so promoting a good holdout sentence into a template cannot silently contaminate training.

For a reject class, accuracy is the wrong headline. Track the two numbers that trade off against each other:

  • OOS recall — off-domain inputs correctly routed to fallback
  • false-fallback rate — in-scope inputs wrongly sent to fallback (the real cost; this is what annoys customers)

A model at 99% on the 16 business intents with 0% OOS recall is worse in production than one a point lower with 85%.

How it was built

Templates → bounded slot fills → sampled surface variants, with the split assigned at step one. Stages that exist because real messages have properties templates don't:

  • Code-mixing — a Banglish→Bengali lexicon flips a random 40–80% subset of words mid-sentence. Latin loanwords (account, balance, transfer, card, OTP, EMI) are deliberately excluded from the lexicon: Bangladeshi users type those in Latin even inside an otherwise-Bengali sentence, and that asymmetry is the pattern worth learning.
  • Phonetic noise — Banglish misspelling is sound-level substitution (bh↔v, sh↔s, ph↔f), dropped vowels (kemonkmon) and word-boundary drift (koto takakototaka), not random character swaps.
  • Fragments — context-free follow-up turns (koto?, kothay, hoyni) where the intent rides on 1–4 words.
  • Glued social openersassalamu alaikum vai amar balance koto, labelled balance_inquiry.
  • Rambling preambles — a sentence of context before the actual question, so the model sees inputs longer than 8 words.

Reproduce with python generate_domain_data.py banking (seeded, deterministic). Adding a template to one intent does not reshuffle any other intent's split assignment.

Limitations and bias

Please read this section before using the dataset as a benchmark.

  • Synthetic. Generated from hand-written templates, not collected from users. It encodes one author's model of how customers write, including its blind spots. A model at 0.71 here is not a model at 0.71 in production.
  • Short inputs. Mean under 5 words. Models trained here will be poorly calibrated on long multi-paragraph messages.
  • Under-represented code-mixing. 353 mx rows (6%) versus a real inbox where code-mixing is far more common than that. It is seasoning here, not a first-class script.
  • Bangladesh-specific. Payment wallets (bKash, Nagad, Rocket), local bank and branch vocabulary, cities, festivals (Eid, Puja), and honorifics (vai, apu) are all local. Indian retail-banking vocabulary is absent entirely.
  • Romanization is not standardized. Banglish has no orthography. The phonetic-variant generator covers a fraction of real spelling space, and its substitution rules are hand-picked rather than learned from data.
  • Label noise on the overlapping boundaries. The tie-breaks above are applied consistently by construction, but they are one defensible reading of genuinely ambiguous cases. Your product may want them drawn elsewhere.
  • No inter-annotator agreement figure, because there was one annotator.
  • No PII — no real account numbers, names, phone numbers or addresses. Account and transaction references are made-up strings from a fixed list.

Intended and out-of-scope uses

Intended: bootstrapping a Bangla/Banglish banking intent classifier before you have logs; benchmarking small CPU models (fastText, distilled transformers) on code-mixed short text.

Not intended: as evidence of production accuracy; as a general Bangla NLP benchmark; for any high-stakes routing (payments, disputes, fraud, legal) without a human in the loop and a calibrated reject threshold. Money movement in particular should never be triggered by this classifier alone.

Citation

@misc{banglabankingintent,
  title  = {BanglaBankingIntent: Bangla / English / Banglish Banking Intent Classification},
  year   = {2026},
  note   = {Synthetic dataset, 17 intents, template-disjoint splits},
  howpublished = {\url{https://huggingface.co/datasets/Badhon/BanglaBankingIntent}}
}

Licensing

CC BY-NC-SA 4.0 (Creative Commons Attribution-NonCommercial-ShareAlike 4.0).

The content is wholly generated from templates written for this repository, so there is no upstream corpus license to inherit. What the terms mean in practice:

  • BY — attribute the source when you use or redistribute it.
  • NCno commercial use. Training a classifier that serves a commercial bank is a commercial use. If this dataset is meant to be deployable inside a business, cc-by-sa-4.0 or apache-2.0 is the licence you want instead.
  • SA — derivatives, including modified or extended versions of the data, must carry the same licence. Whether a model trained on it counts as a derivative work is legally unsettled and jurisdiction-dependent.

Add a LICENSE file containing the full CC BY-NC-SA 4.0 text alongside this card; HuggingFace renders the tag either way, but the file is what makes the grant explicit to anyone who downloads the CSVs on their own.

Downloads last month
25