2025-12-28

1. The Challenge of Digital Communication

Writing on mobile devices and web interfaces often suffers from latency and human error. Our goal was to build a tool that:


  • Reduces Keystrokes: Anticipates user intent before they finish typing.

  • Understands Context: Adapts suggestions based on the topic (Sports, Social, etc.).

  • Preserves Flow: Handles capitalization and grammar naturally.

  • User Friendly: Provides the top three predicted words for the user to just click on to add while typing.


The Solution: An optimized NLP engine that balances massive linguistic data with real-time performance.

2. The Engine Under the Hood

IntelliType uses a Large-Scale N-Gram Back-off Model trained on over 600MB of diverse text data.

  • N-Grams: We utilize Trigrams (3-word sequences) and Bigrams (2-word sequences) to find patterns.

  • Smart Back-off: If a 3-word pattern isn’t found, the app “backs off” to 2-word patterns, and finally to the most frequent 1-word fallbacks.

  • Contextual Boosting: A unique “Topic Engine” scans your recent words and multiplies the scores of relevant vocabulary (e.g., typing “wife” boosts relationship-related words).

  • Dictionary Coverage: The model utilizes a curated vocabulary of 7,000 unigrams, capturing approximately 90% of the lexical instances found in the training corpus. .

  • Small Memory Footprint: a Total of ~120 MB for the application and the n-grams lookup tables loaded.

  • Low Latency Rate : Less than 150 ms average latency (after ~1 sec initial loading of the app and the tables).

3. Quantitative Performance

We evaluated the model using a “hold-out” testing dataset, the app was not trained on

Metric Performance Why it matters
Top-1 Accuracy 13.9% How often the first button is exactly right.
Top-3 Accuracy 28.6% How often the word is among the 3 buttons.
Total Avg. Latency < 150 ms Suggestions appear instantly as you type.
Lookup N-grams Search Speed <100 ms Rael-time response using data.table keys
Dictionary Size 7,000 Words High-frequency core vocabulary.
Memory Footprint ~120 MB High efficiency allows for use on any device.
Disk Space ~14 MB Compressed storage for easy deployment.

4. User Experience & Interface

IntelliType was designed with an “Integrated Approach” to maximize focus and minimize clutter.

  • Type-Ahead Sensing: Suggestions filter in real-time as you type partial words.
  • Case-Sensitivity: If you type “Adam S”, it suggests “Sandler” (Capitalized).
  • Integrated Clear: A sleek, nested reset button allows for instant editing.
  • Live Monitor: A dashboard view shows you exactly which part of the algorithm is making the choice, and what context has been recognized.
  • Top 10 Predicted Words: The dashboard shows the top 10 predicted next-word with their score
  • Context Awareness: Unlike standard keyboards, IntelliType recognizes if you are talking about sports or social event,s and shifts its vocabulary accordingly.
  • Speed: By using data.table indexing in R, the search through 7,000+ unigrams happens in milliseconds.
  • Professional UI: A clean, purple-accented dashboard provides a “Pro” writing environment.


IntelliType…Experience the future of intelligent typing today… Try it at:   IntelliType at Shinyapps.io