Monthly Action Plan

05 Monthly Action Plan (Junior Spring → Senior Winter)

Fatima, the next several months should focus on turning your Somali‑Bantu linguistic materials into a structured, documented dataset and then publishing it publicly with clear technical documentation. The committee emphasized that the strongest signal will come from showing real computational linguistics workflow: organized data, reproducible pipelines, and visible usage after publication. The calendar below sequences that work so that the project is mature before early application deadlines and continues generating evidence of impact afterward. See the Creative Projects section for deeper guidance on the dataset structure and technical approach.

May (Junior Year)
  • Dataset audit and structure plan: inventory all Somali‑Bantu dictionary entries, audio files, and text materials you currently have. Define the dataset schema (tokenized text, aligned audio, speaker metadata, linguistic fields). Target outcome: a written dataset specification.
  • Repository setup: create or organize a GitHub repository that will host the dataset documentation, preprocessing scripts, and future pipeline code. The goal is to begin recording visible technical contributions through commits.
  • Documentation baseline: start a README describing dataset purpose, collection process, and planned structure. This documentation should evolve as the project develops.
June
  • Tokenization and text formatting: convert the dictionary text into a structured machine‑readable format (for example JSON, CSV, or similar). Ensure entries are consistently tokenized and linguistically labeled where possible.
  • Audio alignment preparation: organize audio recordings and link them to their corresponding lexical entries or phrases. Begin building the mapping between text tokens and audio clips.
  • Version‑controlled workflow: commit preprocessing scripts and dataset transformations to GitHub so that the full pipeline is transparent and reproducible.
July
  • Metadata layer creation: add structured metadata fields (speaker identifiers, recording context, dialect information if available, recording quality notes). If this information is missing in places, document those gaps clearly rather than guessing.
  • Pipeline documentation: write clear explanations of how raw materials become the structured dataset—tokenization steps, audio alignment approach, and data cleaning procedures.
  • Evaluation framework: begin outlining how the dataset could be used for computational linguistics tasks (e.g., training or testing models). Document any evaluation scripts you create.
August
  • Dataset completion pass: finalize the first fully structured version of the dataset with tokenized text, aligned audio, and metadata fields in place.
  • Public release preparation: prepare licensing information, dataset documentation, and instructions for other researchers or organizations who might want to use the data.
  • Technical transparency: ensure the GitHub repository clearly shows your contributions through code commits, scripts, dataset structure explanations, and pipeline diagrams.
September (Senior Year begins)
  • Public dataset launch: publish the Somali‑Bantu dataset on a public platform (for example a research dataset repository or a documented GitHub release). Target outcome: a stable public version that others can download.
  • Usage tracking setup: configure ways to track downloads, repository stars, citations, or organizational usage. These metrics will become evidence of real‑world impact.
  • Application integration: update your activities list and project descriptions to reflect the dataset publication and your technical contributions. See §06 Essay Strategy for how this project should appear in essays.
October
  • Early application alignment: if you plan to apply Early Action or similar early timelines (for example MIT or the University of Minnesota–Twin Cities), ensure the dataset and documentation are clearly linked in your application materials.
  • Technical contribution log: continue adding commits for improvements—cleaning scripts, evaluation tools, or additional preprocessing utilities.
  • Impact monitoring: begin recording early usage signals such as downloads, forks, citations, or inquiries from organizations.
November
  • Iteration release: publish an updated dataset version if improvements or corrections have been made. Maintain version history so others can cite the dataset reliably.
  • Usage documentation: track metrics such as downloads, repository engagement, or external use. Maintain a simple log that records these indicators over time.
  • Technical write‑up expansion: strengthen repository documentation describing the training pipelines, evaluation scripts, or preprocessing methods associated with the dataset.
December
  • Project impact update: review and summarize dataset usage metrics collected since the release (downloads, citations, or institutional use if visible).
  • Pipeline refinement: continue documenting scripts used for dataset preparation, evaluation, or model experimentation so the project reflects authentic computational linguistics work.
  • Application updates: incorporate the dataset publication and usage indicators into regular decision applications such as West Chester University of Pennsylvania if applicable.
January
  • Impact tracking continuation: update your dataset metrics log and maintain the repository with small improvements or corrections.
  • Technical portfolio maintenance: ensure your GitHub repository clearly shows the full workflow—data structuring, scripts, model experimentation, and documentation.
  • Long‑term visibility: consider additional outreach or documentation improvements that help researchers or organizations discover and use the dataset.

Across all months, the most important ongoing habit is maintaining clear, visible technical documentation. Every dataset transformation, preprocessing script, model experiment, or evaluation tool should be tracked through GitHub commits and explained in the repository documentation. That transparency allows admissions readers—and potential collaborators—to see the depth of your computational linguistics work rather than just the final dataset.

This timeline ensures that by early application deadlines the project already exists publicly, and by regular decision deadlines it can show measurable usage and sustained technical development.

← Testing Strategy Success Stories →