 Courses from Hannah Davis

Natural is a general natural language facility for nodejs. Tokenizing, stemming, classification, phonetics, tf-idf, WordNet, string similarity, and some inflections are currently supported.

natural

In this lesson, we’ll see how to use Natural’s probabilistic spell-checker, which uses the trie data structure. 

Include spell-check in text projects using Natural

In this lesson, we’ll take a look at Natural’s phonetics feature. We’ll learn how to check whether two words sound alike, looking at both the SoundEx and Metaphone algorithms.

Check if words sound alike using Natural

We will learn how to compare how similar two strings are to each other, examining three algorithms: Jaro-Winkler, Levenshtein, and Dice’s Coefficient.  

You should note that none of these algorithms are inherently better than the others. Instead, it's important to choose the one that best fits your text data. 

Compare similarity of strings through string distance in Natural

In this lesson, we will learn how to train a Naive Bayes classifier or a Logistic Regression classifier - basic machine learning algorithms - in order to classify text into categories.

Classify text into categories with machine learning in Natural

In this lesson, we will learn how to train a Naive Bayes classifier and a Logistic Regression classifier - basic machine learning algorithms - on JSON text data, and classify it into categories.

While this dataset is still considered a small dataset -- only a couple hundred points of data -- we'll start to get better results.

The general rule is that Logistic Regression will work better than Naive Bayes, but only if there is enough data. Since this is still a pretty small dataset, Naive Bayes works better here. Generally, Logistic Regression takes longer to train as well. 

This uses data from Ana Cachopo: http://ana.cachopo.org/datasets-for-single-label-text-categorization

Classify JSON text data with machine learning in Natural

Tries are a data structure that provide an efficient way to search for the existence of a word or phrase in a body of text, or to search by prefix. 

Search more efficiently with tries using Natural

Tf-idf, or term frequency-inverse document frequency, is a statistic that indicates how important a word is to the entire document. This lesson will explain term frequency and inverse document frequency, and show how we can use tf-idf to identify the most relevant words in a body of text. 

Identify the most important words in a document using tf-idf in Natural

This lesson introduces WordNet, which is an important resource in natural language processing. With WordNet, we can look up a word’s definition, or find its synonyms.

Find a word’s definition using WordNet in Natural

By this point we've seen that classification can take a long time, and with more data, it would take even longer. Luckily, Natural provides support to save your classifiers. In this lesson, we will learn how to save a classifier and load it into a new project in order to classify new data.

Using machine learning classifiers in a new project

We will learn about “stemming,” the process of finding the root of words, often in order to group words by a common base root. We will look at the Porter and Lancaster Stemmers, briefly touch on Natural’s support for Russian and Spanish stemmers, and introduce the function to stem and tokenize at the same time.

Find the roots of words using stemming in Natural

Inflectors are the modifiers of a word that indicate grammatical categories. While Natural’s coverage of inflectors is not comprehensive, we will show how Natural can pluralize/singularize nouns and count numbers. 

Pluralizing nouns and counting numbers with inflectors in Natural

N-grams are sequences of words, where the 'n' stands for the number of words in the sequence. In this lesson, we will see how to find bigrams (2-grams), trigrams (3-grams), and any other length n-gram in a body of text. 

Find sequences of words (n-grams) using Natural

An important component of many natural language processing projects is being able to identify the grammar of a piece of text. We’ll learn how to do that with Natural’s parts of speech (POS) tagger.  

There are many tags, and it's worth looking them up online (search "POS tag symbols") to become familiar with them all. 

The setup of the tagger may seem a little strange, but it allows you to replace the lexicon or the rules with a different lexicon or rule set of your choice. 

Tag parts of speech using Natural

In this course we’ll work through Natural’s API for natural language processing in JavaScript. We’ll look at how to process text: learning how to break up language strings, find the word roots, work with inflectors, find sequences of words, and tag parts of speech. We’ll learn how to find important stats about a body of text: how to compare strings, how to classify text with machine learning, how to use the tf-idf tool to find relevant words. We’ll look at some of the extra tools Natural gives us, including the dictionary/thesaurus of WordNet, a phonetics comparer that lets us see if two words sound the same, and a spellcheck feature. We’ll also look at tries and digraphs, two data structures that help us better analyze bodies of text.

Natural Language Processing in JavaScript with Natural

A part of Natural Language Processing (NLP) is processing text by “tokenizing” language strings. This means we can break up a string of text into parts by word, sentence, etc. In this lesson, we will use the `natural` library to tokenize a string. First, we will break the string into words using `WordTokenizer`, `WordPunctTokenizer`, and `TreebankWordTokenizer`. Then we will break the string into sentences using `RegexpTokenizer`.

Break up language strings into parts using Natural

Mock Service Worker is an API mocking library that allows you to write client-agnostic mocks and reuse them across any frameworks, tools, and environments.



Mocking WebSocket APIs with Mock Service Worker


When attempting to load the same module twice in JavaScript you'll hit a cache and code won't re-run. In scenarios where you actually do want to have state in your modules, you'll have to use a cache-busting technique by passing a query parameter to a dynamic import. This lesson walks through how to add a query parameter, how to add the appropriate types for loading dynamic modules in TypeScript, and when these techniques are valuable.

Import the Same JavaScript Module Multiple Times with Cache Busting



The include keyword in GitHub Actions empowers you to add custom variables to your matrix strategy, providing flexibility beyond standard configurations. In this video, we'll explore how include allows you to define specific combinations of variables, enhancing control over your workflow runs. You'll see how this feature enables you to tailor jobs for unique environments or scenarios not covered by the default matrix options. By leveraging include, you can efficiently manage complex build processes with greater precision. This technique helps you streamline your CI/CD pipeline, making it more adaptable to your project's evolving needs.

Use GitHub Actions Matrix Include for Complex Job Configurations

While running `zod` schema validation agains domain-rich objects, the error message saying that _a string is not a number_ might not be really useful.

Especially, when you need to provide a meaningful feeback to the user, e.g. after they have submitted a form with lots of data.

`zod` allows to customize messages on a very granular level. We can also extract repeatable customization to `zod` error maps.

Finally, in order to process an error message further (e.g. display it to the user), we need to handle the `zod` validation errors effectively. Especially, when a big object validation has failed and we could receive a handful of errors.

That's what we learn in this lesson. 🔥

Customize zod Error Messages and Handle zod Errors Effectively

customize zod validation messages, customize zod schema property messages, reuse error maps across different schemas, navigate through zod error objects

You might want to run `zod` schema checks in different styles.

Sometimes you prefer to conditionally throw the error - that's what `parse` does. But another time you'd prefer to receive a `boolean` property that determines whether the validation was successful or not - and perform further actions accordingly.

Additionally, the error-throwing API of `parse` works extremely well with [TypeScript assertion functions](https://www.typescriptlang.org/docs/handbook/release-notes/typescript-3-7.html#assertion-functions). Their combination is just... perfect! 👌

That's what we learn in this lesson.

Use parse and safeParse Runtime Type-checks with zod

use zod parse function to conditionally throw ZodError, use zod safeParse to get a boolean value and run typescript type guard over success property

In this lesson we learn that TypeScript cannot guarantee correct shapes of responses. That's due to TS performing checks in compile-time, while the response will be availbale in runtime - when TS doesn't exist anymore.

In some situations, such as HTTP requests, runtime type checks such as `zod` are essential for providing quick feedback around data-related errors.

Understand where are the possible mistakes to make while using TypeScript with HTTP calls. Use `fetch` and `axios` along with `zod` schema checks.

Also, note that similar mistakes or mistakes could happen while using web storage (e.g. `localStorage`)

Validate HTTP API Response with zod, fetch and axios

use zod as an HTTP response validation tool along with fetch and axios in order to make sure that the response has expected data structure

In this lesson we learn that the `zod` schemas become the single source of truth.

We shall not update TypeScript types manually, since they are generated automatically from the schema.

Same with validation functions.

Whenever the schema changes, all the rest will update accordingly. It's an important mind shift.

Treat zod Schemas as Single Source of Truth

Zod Schemas are the single source of truth for both typescript types, and interfaces - and runtime validation functions. Schemas should be updated first.

`zod` is powerful, but writing the schemas manually for existing JSON objects, especially big ones, can be time-consuming.

But that's what `quicktype` is for! You can get 80% work done instantly! 🥳

Depending on your need - whether you want to automate the process, or do it just one time, you can:
- use an online app
- use an IDE extension
- run a CLI script
- generate a script to be executed within CI/CD environment/tooling

And this is what we learn in this lesson! 🔥🔥🔥

Get started with zod quickly!

Generate zod Schemas Automatically using Quicktype App, Extensions and Scripts

Use quicktype app, extension or scripts to automatically generate zod schemas from existing json files. Save time and get started with zod quickly!

`zod` is a powerful library allowing to run runtime type-checks.

It's complementary to TypeScript, which does type checks in compile time.

In this lesson we create few schemas that represent primitive type values, along with `zod` utilities which make these more precise. We also create object schemas and nest one inside another.

All schemas are being validated against example values. We analyze how to interpret error messages in case of failures.

Perform Runtime Type-checks on Primitives and Objects with zod

Create zod schemas to validate primitive type and object type values in runtime. Analyze the errors to see how a value doesn't correspond to schema expectations


Using Cursor Chat, you can easily reference an entire directory of files and ask it to generate anything based on those files. This opens up the possibility of using your current code as a reference guide to create new code that follows existing patterns. With this approach, the result can be a polished prompt you can use with any AI to generate consistent and reliable output.

Focus less on crafting the perfect prompt upfront - what matters most is testing and iterating based on the AI's responses. The more results you see from the prompt, the more you can reinforce the good results and prune out the bad results until your prompt is polished enough that you can rely on the prompt to give you consistent quality results.


Hannah Davis

Search Results