We have been making adjustments to our infrastructure in the past few work days. Submissions may not complete successfully. If you are experiencing problems, please reach out to us on babelmachine [at] poltextlab [dot] com. We thank you for your patience and we apologize for the inconvenience.
Introduction
The Aspect-Based Sentiment Analysis (ABSA) Babel Machine is a dedicated three-label sentiment classifier that predicts Negative, Neutral, and Positive classes (0 = Negative, 1 = Neutral, 2 = Positive). Unlike general sentiment models that classify entire sentences, the ABSA model focuses on specific aspect terms within a sentence and determines the sentiment expressed towards each of them.
The model is built on the BERT-SPC (Sentence Pair Classification) architecture, which jointly considers the target aspect term and its surrounding sentence context. This design allows the model to capture fine-grained sentiment directed at individual aspects, even when multiple aspects appear in the same sentence.
The ABSA Babel Machine processes text at the sentence level. In the output file, each sentence will be listed as many times as it contains aspect terms, with each row displaying the given aspect and the corresponding sentiment prediction. For optimal performance, we recommend providing input that is already segmented into sentences. It should be noted that currently only Named Entities (PER, LOC, ORG, MISC) are considered as aspect terms due to automatic processing constraints.
Current limitation: Since most of the training data contains sentences where all named entities share the same sentiment label, the model performs best in such scenarios. Cases where different named entities within the same sentence are associated with sharply contrasting sentiments are relatively rare in the data; therefore, the model may struggle with these. As a result, it can be challenging to find demo examples where multiple aspect terms receive correctly differentiated predictions. The model tends to predict more reliably when only a single named entity is present in the sentence.
The model currently supports ENGLISH data. You can upload your datasets here for automated aspect-based sentiment coding. If you wish to submit multiple datasets one after another, please wait 5–10 minutes between your submissions. There are two possibilities for upload: pre-coded datasets or non-coded datasets. The explanation of the form and the dataset requirement is available here.
The upload requires you to fill out the following form with metadata about the dataset. Please upload your dataset, and if a pre-coded dataset is available, please attach the codebook used with it.
The non-coded datasets should contain an id and a text column. The column names must be in row 1. You are free to add supplementary variables to the dataset beyond the compulsory ones in the columns following them.
You are free to add supplementary variables to the dataset beyond the compulsory ones in the columns following them. Automatic processing requires following these rules.
If the files you would like to upload are larger than 750 MB, please split your submission into multiple files and submit them one by one. We encourage splitting into smaller file sizes (i.e., 100-300 MB) as the system can more reliably handle smaller files and process them faster than larger files. We recommend waiting some time between each submission, as too many files at once will overload our systems. Please keep in mind that the processing time also depends on the unit of observation as well; a full-text document per row will take longer to process than a sentence per row.
Please use our contact form to let us know if you have any questions or feedback regarding the Babel Machine. Please keep in mind that we can only respond to you on Hungarian business days.
Submit a dataset:
This project was supported by the Ministry of Innovation and Technology NRDI Office within the RRF-2.3.1-21-2022-00004 Artificial Intelligence National Laboratory project; the V-Shift Momentum Project of the Hungarian Academy of Sciences; Miklós Sebők's Excellence project (identifier: 151324), which is funded by the Hungarian National Research, Development and Innovation Office's National Research Excellence Programme; and received additional funding from the European Union's Horizon 2020 program under grant agreement no 101008468. We also thank the Babel Machine project and HUN-REN Cloud (Héder et al. 2022; https://science-cloud.hu) for their support. We used the machine learning service of the Slices RI infrastructure (https://www.slices-ri.eu/).
HOW TO CITE: If you use the Babel Machine for your work or research, please cite this paper:
Sebők, M., Máté, Á., Ring, O., Kovács, V., & Lehoczki, R. (2025). Leveraging Open Large Language Models for Multilingual Policy Topic Classification:
The Babel Machine Approach. Social Science Computer Review, 43(2), 295–317. https://doi.org/10.1177/08944393241259434
GDPR Compliance Statement
- Nature of the Uploaded Data: The files uploaded by users to the tool do not contain personal data as defined in Article 4(1) of the GDPR, which specifies personal data as "any information relating to an identified or identifiable natural person ('data subject')".
- Data Process: The files submitted to our tool are stored in a secure cloud environment to allow processing and generation of the output (the coded CSV file). Personal data provided in connection with the file upload—such as the submitter's name, email address, and similar details—are used exclusively for the purpose of sending the coded files back to the user and identifying the organisation of our users. This processing is conducted in compliance with the purpose limitation principle (Article 5(1)(b)) and the data minimisation principle (Article 5(1)(c)) of the GDPR. By submitting the files, the user consents to this data processing, which is strictly limited to returning the results and identifying the file owner. The personal data is stored securely and retained solely for these purposes. In accordance with Article 17 of the GDPR (Right to Erasure, or "Right to be Forgotten"), users may request the deletion of their personal data at any time. Such requests will be processed promptly, and all related personal data will be permanently deleted from our systems.
- Training Purposes: We do not use personal data to train machine learning models or perform any other type of analysis. When submitting files, the submitter must declare that the uploaded CSV files do not contain any personal data, as stated in the consent agreement. This approach aligns with the purpose limitation principle (Article 5(1)(b)) of the GDPR, which requires data to be collected for "specified, explicit, and legitimate purposes" and not further processed in a manner incompatible with those purposes.
- Google Cloud Platform Compliance: The files submitted to our tool are stored in a secure cloud environment provided by Google Cloud Platform, with configurations ensuring that all processing occurs on servers located within the European Union (EU). This guarantees compliance with GDPR requirements related to data residency and cross-border data transfers. The use of Google Cloud Platform as our processing environment ensures high levels of data security and compliance with GDPR, including the application of the Standard Contractual Clauses (SCCs) for any necessary data transfers. Google Cloud's infrastructure is certified under internationally recognised standards, such as ISO 27001, ISO 27017, and ISO 27018, further ensuring the security and confidentiality of uploaded data.