Datasets:
license: mit
task_categories:
- translation
language:
- id
size_categories:
- 1K<n<10K
Dataset Details
Num2Txt is a dataset designed for converting numerical values into their textual representations in Indonesian. The dataset was generated using a tristage random sampling method:
Randomly select one of 7 groups (units, tens, hundreds, thousands, millions, billions, trillions). Randomly pick a number from the selected group. Skip duplicates to ensure unique entries. This method ensures proportional representation, diversity, and no duplicates across 10,000 entries.
This dataset is particularly suited for natural language processing (NLP), text-to-speech (TTS), and other machine learning applications.
Bahasa Indonesia
Tentang Dataset
Num2Txt adalah dataset yang dirancang untuk mengubah angka menjadi penyebutan teks dalam bahasa Indonesia. Dataset ini mencakup angka acak dari rentang 0 hingga 999 triliun, dengan 10.000 entri yang disusun secara seimbang untuk mewakili angka kecil dan besar secara proporsional. Dataset ini cocok untuk berbagai aplikasi seperti pemrosesan bahasa alami (NLP), teks ke suara (TTS), dan pembelajaran mesin lainnya.
- Curated by: Navillera AI Projec
- Language(s) (NLP): Bahasa Indonesia
- License: MIT License
Uses
NLP: Training models for number-to-text conversion. TTS: Enhancing text-to-speech systems for Indonesian. Language learning tools: Assisting learners in mastering number expressions.
Dataset Structure
The dataset includes two primary fields:
number: Numeric value (integer). text: Textual representation of the number in Indonesian (string). [More Information Needed]
Dataset Creation
Source Data
Data Collection and Processing
Data Generation Process The dataset was generated using code that randomly selects numbers within the range of 0 to 999 trillion (0–999,000,000,000,000). The dataset was generated using a tristage random sampling method:
Stage 1 - Group Selection: The number range (0–999 trillion) was divided into 7 groups based on magnitude:
Units (0–9) Tens (10–99) Hundreds (100–999) Thousands (1,000–99,999) Millions (100,000–99,999,999) Billions (1,000,000,000–999,999,999,999) Trillions (1,000,000,000,000–999,000,000,000,000) One group is selected randomly for each entry based on proportional allocation (e.g., units have fewer allocated positions than billions).
Stage 2 - Member Selection: Once a group is selected, a random number is chosen from that group.
Stage 3 - Duplicate Check: If the chosen number already exists in the dataset, it is skipped, and the process is repeated to ensure no duplicates.
This method ensures:
Proportional representation of numbers across groups. Diversity of numbers within the dataset. No duplicates across the 10,000 entries.
####Bahasa Indonesia Proses Generasi Data Dataset ini dihasilkan menggunakan metode Tristage Random Sampling:
Tahap 1 - Pemilihan Kelompok: Rentang angka (0–999 triliun) dibagi menjadi 7 kelompok berdasarkan besaran:
Satuan (0–9) Puluhan (10–99) Ratusan (100–999) Ribuan (1.000–99.999) Jutaan (100.000–99.999.999) Milyaran (1.000.000.000–999.999.999.999) Triliunan (1.000.000.000.000–999.000.000.000.000) Untuk setiap entri, satu kelompok dipilih secara acak berdasarkan alokasi proporsi tertentu (misalnya, kelompok satuan memiliki posisi lebih sedikit dibandingkan milyaran).
Tahap 2 - Pemilihan Anggota: Setelah kelompok dipilih, satu angka dipilih secara acak dari kelompok tersebut.
Tahap 3 - Pemeriksaan Duplikasi: Jika angka yang dipilih sudah ada dalam dataset, maka angka tersebut dilewati, dan proses diulang untuk memastikan tidak ada duplikasi.
Metode ini memastikan:
Representasi angka yang proporsional antar kelompok. Keragaman angka dalam dataset. Tidak ada duplikasi pada 10.000 entri dataset. Dengan deskripsi ini, proses Tristage Random Sampling menjadi jelas dan mudah dipahami. Jika perlu tambahan detail, beri tahu saya! 😊
Bias, Risks, and Limitations
[More Information Needed]
Recommendations
Users should be made aware of the risks, biases and limitations of the dataset. More information needed for further recommendations.
Citation [optional]
BibTeX:
[More Information Needed]
APA:
[More Information Needed]
Glossary [optional]
[More Information Needed]
More Information [optional]
[More Information Needed]
Dataset Card Authors [optional]
[More Information Needed]
Dataset Card Contact
[More Information Needed]