The Lighthouse · Signature Series

How Machines Work with Language: Classical NLP for Sentiment, Search and Translation

Signature Course No. 338 — A Signature AI Socratic Coaching Course.

No experience needed — your coach starts where you are.

Preview This Course — Free
An agentic, mastery-based AI learning platform.

A New Kind of Learning

Socratic coaching, delivered by an agentic AI coach.

Not a course you watch. An AI coach you work with.

The Lighthouse Coach begins by discovering how you think about the subject, teaches what is missing, and gives you realistic practice before asking you to prove that you can use it. You may test out of concepts you already understand, while capstones turn the learning into a practical plan, decision, conversation, or other real-world result.

The curriculum is deliberately authored; the AI delivers it through coaching. It is not an AI randomly generating lessons.

Meet The Lighthouse Coach

A coach that teaches by asking — not by lecturing

The Lighthouse Coach is an AI Socratic coach built on Cognosco Learning's How Machines Work with Language: Classical NLP for Sentiment, Search and Translation course, in the AI at Work category. It does not play videos at you. It holds a conversation: it asks you a question, listens to your answer, names the misconception it hears, affirms what you got right, and asks the next question — the way a great coach works in the room.

A machine cannot reason about a sentence until the sentence has been turned, honestly, into numbers. Before a machine can judge whether a review is happy, find the one help article that answers a question, or carry a sentence across a language, it must turn text — the least numerical material in the world — into numbers it can add, compare, and weigh. This course teaches that conversion and everything built on it, in the order the field itself built it.

It starts with the cutting. You will split text into tokens and build a vocabulary under rules you can state; count words into vectors and weight the words that carry meaning; score sentiment first with a word list you can audit and then with classifiers fitted to labeled examples whose arithmetic you can reproduce; and buy word order back — tags, n-gram probabilities, and a hidden Markov model whose best path you decode yourself, grid by grid.

It works in two movements. First, Module 1 — From Raw Text to Counted Evidence — work through Concepts 1–9, finishing with The Classical Triage Pack; then, Module 2 — Sequences, Meaning, and Translation — work through Concepts 10–18, finishing with The Multilingual Language Pack.

Both capstones produce real artifacts — a complete classical triage pack for Kestrelwick Outfitters' 8,400-message monthly stream: the token standard, TF-IDF search over its 180-article help library, a three-layer sentiment triage with worked verdicts, sequence machinery with a complete Viterbi decoding, and an audit page with its limits named; and a multilingual language pack that carries the same machinery across a language boundary — its search, alignment, translation, and evaluation design, defended in writing.

The honesty standard is part of the course. This course teaches the foundations underneath modern language models, not applications built on top of them: there is no prompt-writing instruction here and no hosted language-model service to call, because neither teaches you how the machinery works. Classical methods are transparent, cheap, and often surprisingly strong — and their limits are named concept by concept, and the course makes no promise that completing it qualifies you for any role, credential, or result.

The course is self-paced and always on — roughly 30 minutes per concept, about 9 hours in full, taken in as many sittings as you like. You can type your answers, or talk out loud: the Voice Coach speaks its questions and listens to your spoken replies. Learners who already know the material can test out concept by concept.

How Machines Work with Language: Classical NLP for Sentiment, Search and Translation is part of The Lighthouse Signature Series — the online counterpart to the one-day, in-person learning Cognosco Learning has delivered since 2002.

Course Facts

Concepts18 · 2 modules
Capstones2
Time per concept≈ 30 minutes
Full course≈ 9 hours, self-paced
FormatOnline · AI voice coach
CertificateOn completion
VerificationVerifiable online
Introductory price$195

One full concept, coached end to end — free: verify your email and we’ll send you a link to start.

Already enrolled? Enter the course →

The Method

Diagnose → Learn & Practice → Prove → Capstone

Every learner moves through the same discipline a Cognosco instructor would insist on in the room.

1

Diagnose

The Coach starts where the course starts: with the cutting. Take a recent customer message and say which words in it your current tools treat as different words though a human reads them as the same — and which the same though a human reads them as different; name the smallest word whose removal would change a message's meaning in your work; and say what actually happens today when your systems meet a word they have never seen — a shared unknown token, a silent drop, or a failure you cannot name. Every concept opens with three questions that locate where you actually stand, and your answers shape how deeply the Coach works the concept with you. Everyone begins with Module 1 — no skipping the foundations you only think you have.

2

Learn & Practice

Work each concept in conversation with the Coach — spoken or written — then practice it against realistic situations until it holds up.

3

Prove

Every concept ends in proof: answer the check cold and demonstrate the skill. Know it already? Test out and take credit immediately.

4

Capstone

Finish with two capstones — The Classical Triage Pack for Module 1 and The Multilingual Language Pack for Module 2 — that pull the whole course together, and earn the certificate.

The Course Content

What You'll Learn

One machinery, three views of it. The three applications in the title — sentiment, search, and translation — are not three separate subjects here. They are three views of one machinery, and by the second module you will be moving parts between them on purpose: the weighting that ranks a help article is the weighting that prices a word in a translation table.

Module 1 turns text into counted evidence. Text into Tokens; Counting Words; Weighting What Matters; Sentiment by Hand; Learning from Examples; Drawing the Line; Tagging Every Word; Predicting the Next Word; and Hidden States — the decisions that fix what can ever be counted, matched, or learned, each one computed by hand on Kestrelwick Outfitters' standing figures.

Module 2 teaches the evidence to keep its shape. Finding Names and Edges; Grammar in Trees; Words as Vectors; Finding Themes; Searching on Purpose; Words Across Languages; Sentences In, Sentences Out; Where to Look; and Judging the Machine — order, structure, meaning, and finally another language, closing with the discipline of judging any of it from its errors.

And everything is worked by hand, at small scale. You will count words in three short documents and compute the weights yourself, multiply the probabilities of a two-word hidden Markov model on paper and watch the winning path win, and divide attention scores into weights that sum to one — because a learner who has only watched large systems produce answers owns vocabulary, not understanding.

What you will learn

  • Turn raw text into tokens, vocabularies, and count vectors, and weight words by how much meaning they carry
  • Score and classify sentiment — with a lexicon you can audit, and with Naive Bayes and logistic classifiers whose arithmetic you can reproduce
  • Model word order with n-gram probabilities, part-of-speech tags, and hidden Markov models decoded step by step
  • Represent meaning geometrically — word embeddings, document similarity, and themes — and say exactly what the geometry does and does not capture
  • Build and judge search: ranking, the BM25 idea, and precision and recall computed from a judged set
  • Carry meaning across languages — word alignment, phrase-based translation, sequence models, and attention — and evaluate any language system from its errors
What's Included

Everything in the course

  • All 18 concepts of How Machines Work with Language: Classical NLP for Sentiment, Search and Translation, in 2 modules
  • A starting-point check first — the Coach finds where you actually stand before teaching
  • Learn, Practice, Prove for every concept — no passive video watching
  • A test-out on every concept: prove you know it and move on
  • The Voice Coach: talk with the Coach out loud — it speaks, listens, and replies
  • A personal Learner Record of your work across the course
  • Two capstone exercises: The Classical Triage Pack and The Multilingual Language Pack
  • A printable session summary whenever you pause
  • The Lighthouse Coach completion certificate (Signature Series)
  • Your progress is saved as you go — stop anytime, resume where you left off
Who It's For

Built for the people who work with text and must defend the numbers

Builders and analysts — who work with text data — reviews, messages, tickets, documents — and want methods they can inspect, explain, and defend, not a black box they rent; and technical staff moving toward language work who can follow a worked count, a small matrix, and a probability product, and want the foundations that modern language systems still stand on.

Product and support owners — who must judge vendor claims about sentiment, search, and translation tools and need to know what the machinery can and cannot do.

The simply curious, at working depth — readers who use language technology every day and want to know, mechanically, what happens between a typed sentence and a machine's answer. This course assumes you can add and multiply fractions and decimals and follow a small table; it assumes no prior programming and no prior linguistics.

Prefer to learn live and in person? The same material anchors Cognosco's one-day programs — browse the in-person catalog.

Enroll

Start with Text into Tokens

$195 — introductory price. One payment. No subscription, no password to manage.

Here is exactly what happens when you enroll:

  • Pay securely with Stripe — you are taken straight into the course
  • Create your learner profile and begin the Module 1 starting-point check
  • Save the access key you are given — it restores your enrollment on any device
  • Learn at your pace; your progress is stored in your browser on your device

For the full voice experience, use Chrome and allow the microphone when asked. Typing your answers works everywhere, all the time. Questions before you enroll? Email hello@cognoscolearning.com.

The Lighthouse Coach — How Machines Work with Language: Classical NLP for Sentiment, Search and Translation$195.00
Total · introductory price, one-time$195.00

Secure checkout by Stripe. The moment your payment completes, you are taken straight into the course to create your learner profile and begin — no account password to manage. You will receive an access key to save; it restores your enrollment on any device.

Good Questions

Before you enroll

What is a Socratic coaching session?

A conversation, not a video. Your coach asks a question; you answer — out loud by voice, or by typing — and the coach responds to what you said: naming the misconception it hears, affirming what you got right, and asking the next question. You advance only when you have mastered each concept, so a finished course means the learning held.

How long does it take — and how long do I have?

About 30 minutes per concept. Learn in sittings of any length; your progress is saved as you go. You have a full year from purchase to finish, and you can revisit and retake within that year.

What does the certificate verify?

That you completed every concept and both capstones of this course. Anyone can confirm a Cognosco certificate online — enter its certificate ID on our verification page.

What if it isn't for me — what are the refunds?

If you have not started, cancel for a full refund within seven days of purchase. Once you have started, there is no cash refund — stop before finishing and we credit the full amount you paid toward any other course in the library (credit valid one year from your original purchase). Once you complete the course, no refund or credit applies.

The Signature Series Continues

More Signature Courses are coming

How Machines Work with Language: Classical NLP for Sentiment, Search and Translation is Signature Course No. 338. Leave your email and we'll send you news of future Signature Courses and Lighthouse updates as they are announced — nothing else.