Event Data.Gov.in

The purpose of this Hackathon is to engage Indian students, researchers, and innovators in developing advanced, data-driven AI and ML solutions based on given data set. Participants will have access to a comprehensive data set containing approximately 900,000 records, each with around 21 attributes and target variables. This data is anonymized, meticulously labeled, and includes training, testing, and a non-validated subset reserved specifically for final evaluations by the GSTN.

Participants are encouraged to use this dataset to design and implement innovative artificial intelligence (AI) and machine learning (ML) algorithms to tackle the stated challenge.

Additionally, this initiative aims to foster collaboration between academia and industry professionals, driving the development of effective and insightful solutions that strengthen the GST analytics framework.

PARTICIPATION

Indian students or researchers associated with educational institutions, or working professionals associated with Indian startups and companies can participate in the Hackathon. The participant must be the citizen of India.

LOGIN AND REGISTRATION

All participants must register at Janparichay. A registered user can directly login at https://event.data.gov.in and submit required details to participate in the Hackathon. It is expected that the participants would submit accurate and up-to-date details and they have to confirm this before submission.

STRUCTURE OF THE HACKATHON

• The Hackathon would be organised as an online event with processes for registration of participants, accessing the datasets to be utilized for each problem statement, and submission of developed prototypes. There would be an offline event with the shortlist participants for the finale/second round.
• Indian students or researchers associated with educational institutions, or working professionals associated with Indian startups and companies can participate in the Hackathon. The participant must be the citizen of India.
• The participants are expected to form teams of up to five members including at least one team lead. A participant may only register as a member of a single team.
• The Hackathon would take place over 45 days from the start of registration to the final date for submission of developed prototypes.
• Participants would receive a dataset containing 9 lakh records with around 21 attributes each. The data is anonymized and labelled, including trained, validated, and non-validated datasets.
• Before submission of solution prototype, participants have to upload their code in GIT (https://www.github.com) repository and an optional demo/product video on YouTube.
• For online submissions, following required/optional fields are to be shared for evaluation:
  • Idea/Concept
  • Project Description
  • Source Code URL (github.com)
  • Project Report (A maximum of four reports can be added, with a minimum of one required)
  • GitHub Unique Source Code Checksum – Steps to create checksum are mentioned in later steps.
  • Video URL (Optional)

• The evaluation process of the Hackathon would be overseen by a distinguished panel of jury members comprising experts from the fields of machine learning, data science, and tax administration. The jury would rigorously assess each submission based on predefined criteria to ensure a fair and comprehensive evaluation.

Given a dataset D, which consists of:

D_train A matrix of dimension R(m×n) representing the training data.

D_test A matrix of dimension R(m1×n) representing the test data.

We have also provided corresponding target variable Ytrain matrix dimension of R(m×1) and 

Ytest   with matrix dimension of R(m1×1).

The objective is to construct a predictive model Fθ(X)→ Y_pred that accurately estimates the target variable Y{i} for new, unseen inputs X{i}

Steps:

1. Model Construction:

Define a predictive function Fθ(X) parameterized by θ that maps input features X to predicted outputs Y_pred.

The model Fθ(X) should be designed to capture the relationship between the input features and the target variable effectively.

Optimize the model parameters θ by minimizing a loss function L(Y,Fθ(X)) using the training data D_train

Consider incorporating feature transformations, feature engineering, or feature selection to enhance the model’s predictive performance.

Apply the learned model Fθ *(X) (with optimized parameters 𝜃∗) to the test data D_test to generate predictions Y_pred for each input Xj∈{X1,X2,…,Xm1}.

            Evaluate the model’s performance by calculating accuracy or other relevant metrics M on the test predictions Y_{pred_test}.

Refine the model by iteratively adjusting θ or modifying  Fθ(X) to improve performance on the chosen evaluation metrics M.

Present the predicted outputs Y_{pred_test} along with a detailed report that includes:

• • The modeling approach employed(Properly commented Codes, supporting citations etc).
  • The metrics used for evaluation.
  • Key performance indicators as per the defined metrics for the hackathon.

** Kindly refer ‘Submission and Expectation’ page before submitting your solutions.

TECH STACK FOR BUILDING AI/ML BASED ALGORITHM

• Participants are encouraged to innovate by developing their unique functions (f(x)) to tackle the given challenge.
• Participants have the liberty to utilize any tech stack of their preference for model development. This flexibility allows them to harness the tools and technologies they are most adept at, facilitating the creation of effective and inventive solutions and deriving the mathematical function for this Hackathon.
• Participants are encouraged to explore and experiment with diverse ensemble techniques, blending different machine learning algorithms to enhance performance and attain optimal results on test data.

PRIZES

The Hackathon offers significant prizes for the top-performing teams, and these are:

1. First Prize: Rs. 25 lakhs
2. Second Prize: Rs. 12 lakhs
3. Third Prize: Rs. 7 lakhs
4. Special Prize of Rs. 5 lakhs for All-Women Teams (in addition to the top three prizes)

• Prizes would only be awarded if the model created meets the jury’s satisfaction of usability of the designed solution as a viable product.
• Consolation prizes of Rs. 3 lakh, Rs. 2 lakh, Rs. 1.5 lakh and Rs. 1 lakh would be given in lieu of announced prizes, if the jury does not find any model provide perfect solution of the problem statement.

CONSOLATION PRIZES

* Note that the prizes declared are for selection after the second round and not the initial stage.

TERMS AND CONDITIONS FOR GST ANALYTICS HACKATHON

These terms and conditions govern the Online Hackathon on GST Analytics Hackathon. By registering and participating in the event, one is deemed to have accepted terms and conditions mentioned below as well as the terms of use of the OGD Platform India.

GENERAL TERMS AND CONDITIONS

Please read these Terms and Conditions carefully as they apply for the Hackathon. To be eligible to participate and declared as shortlisted or winners in the Hackathon, the participants must abide by these Terms and Conditions:

1. Participants must adhere to a high standard of behaviour and professionalism. Harassment, discrimination, and inappropriate behaviour will not be tolerated. Participants must comply with all instructions from organisers.
2. The participating teams may address Problem Statements defined by the GSTN and submit innovative products and services as specified for the Problem Statement.
3. Participants must keep their contact information accurate and up-to-date.
4. Only a single Janparichay/OGD account is permitted for an individual or a team. If more than one account exists for the same candidate or the team, then the candidature of both the team and the individual candidate will automatically result in disqualification.
5. As a part of the submission, the contestant certifies the originality and ownership of the application as detailed/described in the documentation uploaded at the time of submission.
6. The participant(s) must ensure that his/her/their work has not been previously published or awarded.
7. If the participants are acting within the scope of their employment, as an employee, contractor, or agent of another party, then the participants warrant that such party has full knowledge of the actions of the participants and has consented thereto, including the potential receipt of a prize/certificate. The participants’ further warrant that their actions do not violate the employer’s or company’s policies and procedures.
8. The participants will ensure code is free from viruses, malware.
9. The participants will not use this contest to do anything unlawful, misleading, malicious, or discriminatory.
10. Upon submitting the participants agrees that the submitted model shall be the property of the GSTN and the participants grant GSTN the exclusive Intellectual Property Right of ownership.
11. The participant and the participating team agree to take all reasonable measures to protect the confidentiality and avoid unauthorized disclosure of the data provided or use of the submitted model or any other confidential information associated with the model.
12. The winning applications must be maintained in working condition by the contestant(s) for a period of one year. No functional enhancements are expected, but all bugs identified according to the description in the documentation should be fixed immediately on reporting.
13. The models submitted or awarded will become the property of GSTN, including all intellectual property rights to their underlying methodologies and innovations, and the participants shall be deemed to have given their no objection/consent for the same and shall also remain bound by the terms of Non-Disclosure Agreement (NDA) with respect to such work. The participants agree to provide a No Objection Certificate as an author in favour of GSTN for the purposes of IPR registration and ownership rights, as and when required by GSTN.
14. If any participant is determined to have violated the terms of the contest, GSTN/NIC have all rights to disqualify the participant without prior notice.
15. Prizes will be awarded to the winning teams as determined by the Jury. Prizes are non-transferable and no substitution will be made except at the GSTN’s discretion. If the Applications shortlisted do not meet the expectations of the Jury, Jury has the discretion, not to confer an award in one or more categories/ subcategories.
16. The Jury’s decision is final and cannot be challenged.
17. If required, GSTN may change the terms and conditions.
18. The organizers reserve the right in their sole discretion to withdraw participation of any individual/team from the event or reject any submission at any point of time during the process.
19. GSTN shall not be held responsible either directly or indirectly for any damage/s and loss/es, to the participant or the participating team resulting from their participation in the Hackathon. Participants assume all risks associated with their participation.
20. The participants’ personal information shall be used in accordance with the organizers’ privacy policy.
21. By successfully registering on the portal for the Hackathon, it is considered that you agree to the terms and conditions, including the Non-Disclosure Agreement, as stated in the terms and conditions and FAQ section.

NON-DISCLOSURE AGREEMENT

1. The Parties agree to execute this Confidentiality Agreement and be bound by the terms and conditions hereof as a precondition to the proposed negotiations/discussions and agreement between the Parties in relation to the Purpose.
2. “Confidential Information” shall mean all information, know-how, ideas, designs, documents, concepts, technology, commercial knowledge, and other materials of a confidential nature and includes but is not limited to, information of a commercial, technical or financial nature which contains amongst other matters, trade secrets, know-how, patent, Source Codes, IPRs and ancillary information and other proprietary or confidential information, regardless of form, format, media including without limitation electronic, written or oral, and also includes those communicated or obtained through meetings, documents, correspondence or inspection of tangible items, facilities or inspection at any site or place including without limitation:
   • research, development or technical information, confidential and proprietary information on products, intellectual property rights;
   • business plans, operations or systems;
   • details of suppliers;
   • information relating to the officers, directors or employees of GSTN;
   • formulae, IPRs, patterns, compilations, programmes, devices, methods, techniques, or processes, that derive independent economic value, actual or potential, from not being generally known to the public.
3. Except as otherwise provided in this Agreement, the Receiving Party shall keep confidential all information disclosed by GSTN which:
   • is disclosed, communicated or delivered to the Receiving Party in furtherance to the Purpose for which the Parties are entering into negotiations/discussions;
   • comes to the Receiving Party’s knowledge or into the Receiving Party’s possession in connection with negotiations/discussions towards the Purpose.

Notwithstanding whether such Confidential Information is received before or after the date of this Agreement.

4. Except as otherwise provided in this Agreement, the Receiving Party shall not disclose to any other person the status, terms, conditions or other facts concerning the negotiations/discussions as contemplated between the Parties in terms hereof.
5. The Receiving Party shall not use or copy the Confidential Information of GSTN and as both Parties may agree in writing from time to time
6. In the event of the Receiving Party visiting any of the facilities of GSTN, the Receiving Party undertakes that any further Confidential Information which may come to its knowledge as a result of any such visit shall be kept strictly confidential and that any such Confidential Information will not be divulged to any third party and will not be made use of in any way,
7. Except as otherwise provided in this Agreement, the Receiving Party shall not disclose or communicate, cause to be disclosed or communicated or otherwise make available Confidential Information to any third party other than:
   • The Receiving Party’s directors, officers, employees, or representatives to whom disclosure is necessary for the purpose of discussions
   • (each an “Authorised Person”, and collectively, the “Authorised Persons”)
8. The Receiving Party hereby agrees to bind such Authorised Person(s) with similar obligations of confidentiality. In any event, the Receiving Party shall remain liable for any disclosure by the Authorised Person(s) to any other person.
9. The Receiving Party’s obligations hereunder shall not apply to Confidential Information if the same is:
   • in or enters the public domain, other than by breach by the Receiving Party or any of its Authorized Person(s) or
   • known to the Receiving Party on a non-confidential basis prior to disclosure under this Agreement, at the time of first receipt, or thereafter becomes known to the Receiving Party or any of its Authorized Person(s) without similar restrictions from a source other than GSTN, as evidenced by written records, or
   • is or has been developed independently by the Receiving Party without reference to or reliance on GSTN’s Confidential Information.
10. Except as otherwise provided in this Agreement, the Receiving Party may not disclose the Confidential Information of GSTN except if the disclosure is made pursuant to a directive or order of a Government entity or statutory authority or any Judicial or governmental agency provided however that the Receiving Party shall promptly notify GSTN so as to enable GSTN to seek a protective order or other appropriate remedy;
11. The Receiving Party shall exercise no lesser security or degree of care than that Party applies to its own Confidential Information of an equivalent nature, but in any event, not less than the degree of care which a reasonable person with knowledge of the confidential nature of the information would exercise.
12. The Receiving Party acknowledges that any breach of this Agreement by the Receiving Party may cause GSTN irreparable damage for which monetary damages may not be an adequate remedy. Accordingly, in addition to other remedies that may be available, GSTN may seek injunctive relief against such a breach or threatened breach.
13. All written Confidential Information or any part thereof (including, without limitation, information incorporated in computer software or held in electronic storage media) together with any analyses, compilations, studies, reports or other documents or materials prepared by the Receiving Party or on its behalf which reflect or are prepared from any of the Confidential Information provided by GSTN shall be returned to  GSTN or destroyed by the Receiving Party, when requested by  GSTN at any time, or when the Receiving Party’s need for such information has ended or when this Agreement expires or is terminated, whichever is earlier. In the event of destruction, the Receiving Party shall certify in writing to GSTN within thirty (30) days, that such destruction has been accomplished. The Receiving Party shall make no further use of such Confidential Information nor retain such Confidential Information in any form whatsoever.
14. This Agreement shall be effective and perpetually binding from the date of execution hereof.
15. Nothing contained in this Agreement shall be deemed to grant to the Receiving Party either directly or by implication, any right, by license or otherwise, under any patent(s), patent applications, copyrights or other intellectual property rights with respect to any Confidential Information of GSTN nor shall this Agreement grant Receiving Party any rights whatsoever in or to GSTN’s Confidential Information, except the limited right to use and review the Confidential Information as necessary to explore and carry out the proposed Purpose between the Parties.
16. This Agreement is not intended to constitute, create, give effect to, or otherwise recognize a joint venture, partnership or formal business entity of any kind and the rights and obligations of the Parties shall be limited to those expressed set forth herein. Any exchange of Confidential Information under this Agreement shall not be deemed as constituting any offer, acceptance, or promise of any further contract or amendment to any contract which may exist between the Parties. Nothing herein shall be construed as providing for the sharing of profits or losses arising out of the efforts of either or both parties. Each Party shall act as an independent contractor and not as an agent of the other Party for any purpose whatsoever and no Party shall have any authority to bind the other Party.
17. This Agreement contains the entire understanding between the Parties with respect to the safeguarding of said Confidential Information and supersedes all prior communications and understandings with respect thereto. No waiver, alteration, modification, or amendment shall be binding or effective for any purpose whatsoever unless and until reduced to writing and executed by authorized representatives of the Parties.
18. The rights, powers and remedies provided in this Agreement are cumulative and do not exclude the rights or remedies provided by law and equity independently of this Agreement.
19. This Agreement shall be governed and construed in all respects in accordance with the laws of India and exclusively subject to jurisdiction of Courts situated in Delhi.

SUBMISSION & EVALUATION OF MODEL AND ITS IMPACT

• The efficiency and effectiveness of the proposed algorithms would be evaluated against the validation dataset. This rigorous assessment would determine the models’ practical viability and accuracy in real-world applications.
• Submitted models (in first and final rounds) would be evaluated on the following metrics, given that we have provided a binary classification problem. Participants are encouraged to provide the following metrics along with the algorithm (model) at the time of final submission:

1. 1. Accuracy: The proportion of correctly classified instances (both true positives and true negatives) out of the total instances
   2. Precision: The proportion of true positive instances out of the instances predicted as positive.
   3. Recall (Sensitivity or True Positive Rate): The proportion of true positive instances out of the actual positive instances.
   4. F1 Score: The harmonic means of precision and recall, providing a single metric that balances both concerns.
   5. AUC-ROC (Area Under the Receiver Operating Characteristic Curve): AUC represents the degree of separability and measures how well the model distinguishes between classes. ROC is a plot of the true positive rate (Recall) against the false positive rate (1- Specificity).
   6. Confusion Matrix: A table that provides a detailed breakdown of true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN). It helps in visualizing the performance of the classification model.
   7. Other Metrics (Optional): Log Loss and Balanced Accuracy of the model.
   8. Any other additional criteria as decided by jury member.

EXPECTED DELIVERABLES FROM PARTICIPANTS

• Model Code and Documentation:
  1. Clear and well-documented code used to build, train, and test the submitted model.
  2. Explanation of the key methodology and steps taken in model development.

• Model Performance Report:
  1. Evaluation of the model using the provided metrics.
  2. Insights and analysis derived from the model’s predictions.

• Presentation:
  1. Summary of your approach, findings, and recommendations.
  2. Visual aids to support your presentation (graphs, charts, etc.).
  3. Appendices if any

• Citation Report:
  1. Citing all relevant work used, libraries and others along with plagiarism declaration.

• The write-up should follow these formatting guidelines:
  1. Format: PDF
  2. Font: Times New Roman
  3. Font Size: 12 pt
  4. Margins: 1-inch margins on all sides
  5. Line Spacing: 1.5 lines

SUBMIT PROJECT

1. Click on the ‘Submit Project’ on the challenge page. User is redirected on the project submission page.
2. Fill submission form with required and optional fields.
   • Idea/Concept
   • Project Description
   • Source Code URL (github.com)
   • Project Report (A maximum of four reports can be added, with a minimum of one required)
   • GitHub Unique Source Code Checksum – Steps to create checksum are mentioned in later steps.

Note – ZIP in GitHub should have the same checksum as submitted in the submission form. Participant provided checksum should match with the checksum generated at the time of evaluation. Mismatch in these may lead to disqualification.

• • Video URL (Optional)

3. Steps to grant access to your GitHub repository:
   • Go to the main page of your GitHub repository.
   • Click on the ‘Settings’ tab in the menu bar.
   • In the left sidebar, select ‘Collaborators’.
   • Under the ‘Manage Access’ section, click on ‘Add people’.
   • In the text field, search for GSTAnalytics and add it as a collaborator.
4. Steps below to create checksum:
   • Zip compress your complete project.
   • Download the checksum python file from the submission page itself.
   • Install Python on your system. Depending on the system the steps could vary. You can use the following official site for more details – (https://www.python.org/downloads/)
   • Once python installation is complete open terminal.
   • Navigate to the folder directory where the project zip is located.
   • Execute the file checksum.py while giving the file path of the zipped folder as a command line argument. The output will be the Hash of the specific zip file.
   • Example of command when run on Windows 11 with python 3.12.4 installed “python .\checksum.py .\”project_folder_name.zip”
5. Review and modify your registration details -> This provides user a one-time activity to make changes in the registration detail before project submission. Changes will considered final after update.
6. Save As Draft – Participant can save submission details and complete submission process later before deadline. Project submission is not considered complete until it is submitted. Project submission in draft state may lead to disqualification.
7. Submit – On submit, project submission is completed. Mail notification is sent to all team members.
8. Edit Submission -> Participant can submit his project multiple times before the deadline using ‘Edit Submission’ button.
   • Clicking on ‘Edit Submission’ changes project state from ‘Submitted’ to ‘Draft’. Participant must submit project before the deadline and change state to ‘Submitted’.
   • Project with ‘Draft’ state may lead to disqualification

PLAGIARISM AND ETHICS

1. Participants are expected to uphold the highest standards of ethics and integrity throughout the Hackathon.
2. All work submitted must be original and developed by the participant or their team.
3. Plagiarism, or the use of someone else’s work without proper attribution, is strictly prohibited and would result in immediate disqualification.
4. Participants must ensure that their solutions are created from scratch and not copied from existing projects or code repositories.
5. Moreover, the use of any external resources or pre-trained models should be clearly cited, and proper permissions should be obtained where necessary. Adherence to these ethical guidelines ensures a fair and competitive environment for all participants.
6. By registering for this Hackathon, participants are giving an undertaking to adhere to all plagiarism and ethical guidelines set forth by the GSTN.

The evaluation process of the Hackathon would be overseen by a distinguished panel of jury members comprising experts from the fields of machine learning, data science, and tax administration. The jury would rigorously assess each submission based on predefined criteria to ensure a fair and comprehensive evaluation.

JURY COMPOSITION: The jury would tentatively include:

• Senior data scientists with extensive experience in predictive modelling and AI.
• Tax administration experts with deep understanding of fraud detection and related challenges.
• Academic professionals specializing in machine learning and data analytics.
• Representatives from GSTN and NIC with domain-specific expertise.

Jury list would be published shortly.

EVALUATION PROCESS

• Initial Screening: Submissions would undergo an initial screening to ensure compliance with submission guidelines and basic functionality.
• Technical Evaluation: The jury would conduct a detailed technical evaluation of the models with the help of GSTN’s data team, focusing on performance metrics, innovation in approach, and robustness of the solution.
• Practical Usability: Models would be assessed for their practical usability and potential for real-world implementation
• Based on the initial evaluation, 9 to 15 teams would be shortlisted for the second round.
• In second round, teams would refine their models using additional data and insights gained from discussions with SMEs. The final submissions would include a fine-tuned model, a detailed write-up, and a presentation, followed by an interview with the jury in Delhi.

DECISION MAKING

• Prizes would be awarded to the top three teams whose models meet the jury’s satisfaction regarding usability as viable products. A special prize would be awarded to women only team, if any.
• If no solution meets the required standards, consolation prizes would be awarded based on the jury’s discretion.
• The jury’s decision would be final and binding.