A data ontology is a way of linking different forms of data based on different concepts. In the early days of the Internet, data was linked using the HTTP protocol. Today, we can add another layer, an ontology, to define a particular concept and automatically link data points related to that concept.
If you saw this word recently and thought that ontology was new, don’t worry. it is older than the oldest sweater you own. AristotleA 4th century BC Greek philosopher called it “the first philosophy” in his writings. metaphysics.
To be honest, it took a while for the concept to become popular again.German rationalist philosopher Christian Wolf Finally, ontology returned to the mainstream discussion of eighteenth-century philosophy. Since then, philosophers have consistently debated this topic. As well as bar patrons, when it’s late and the booze is flowing. In this article‘Taking concepts from philosophy seminars and taverns, we explore data ontologies and how they work in practice.
Ontology is the relationship with data
In essence, an ontology is a study as it stands. This concept has recently started gaining momentum in the world of computer science through the concept of data ontology.
A data ontology is a way of linking different forms of data based on different concepts. In the early days of the Internet, data was linked using the HTTP protocol. Today, we can add another layer, an ontology, to define a particular concept and automatically link data points related to that concept.
Ontologies and data
In the early 2000s, ontology moved out of the realm of philosophers when it resonated with computer scientists.Veterans like Tim Berners-Lee began to argue for what they called “linked data”. The idea is that data shouldn’t just exist in the form of hypertext documents and hyperlinks between them. Rather, data should be viewed as what it represents (people, places, events, ideas, activities, etc.) and linked in a human-readable way.
This ontology view is a very high-level way of thinking about your data. Perhaps it is not surprising that the tools necessary to put these ideas into practice were not yet available at the time.
But now the Internet is mature. Tools are available. And ontology is experiencing a computer science renaissance.
What is an ontology
Before delving into how ontologies work in the world of data, let’s see what philosophers say about ontologies.
In essence, an ontology is a study as it stands. To make this a little more concrete, an ontology can also be said to be the study of what exists or is real. “Does God exist?” “Are my feelings real?” “What is ‘nothing’ and does it exist?” are all examples of ontological questions.
These are questions you might ask, especially late at night or after a very tough day. .
Philosophers like to make assumptions to explore such issues further. For example, they may assume that God exists. “What is God’s relationship to humans, animals, plants, the sea, the sky?” It also provides information about
What is a data ontology?
See what I did in that subheader. Ontology questions about ontology and data.
Overwhelmed…but let’s get back to business.
Common point The difference between an ontology in philosophy and an ontology in computer science is that it is an attempt to describe everything that exists: entities, ideas, events, and all the relationships between these things.
For example, if you searched for “Paris” 10 years ago, your favorite search engine spewed out a list of links that seemed particularly relevant to your query. Their relevance was determined by the number of times the word “Paris” was mentioned, the number of backlinks to these sites, and many other criteria. SEO expert You can explain it much better than I can.
Fast-forward to today: Tap “Paris” and the search machine knows it’s a city, knows what a city is, and suggests city-related data points such as demographics, neighborhoods, etc. increase. They may also suggest a train route to Paris, as trains are there for ontologists and their relationship with Paris may make them want to visit.
This is the actual ontology.
How Ontologies Make Data Easy to Use
Of course, there are many other ways to organize your data. These include Vocabularies, taxonomies, thesauri, topic maps, logical models, and relational databases. They have the advantage that you can understand them without knowing anything about philosophy.
What makes an ontology special is how flexible they are. If you change a relational database property from integer to float, you must drop the entire column for that property and recreate it using the new property.Worst case you have to recreate the whole data set Because it is not always possible to add new columns relational databaseIt’s a mess!
However, with an ontology, changing a property is as easy as changing the semantics that underpin it. It might sound complicated, but it’s actually as simple as redefining the column that holds this property. The original data set is never lost, nor are the links and indexes that deal with it.
Data ontology example
Specific examples include: contract data setIf you didn’t know anything about the ontology, you could put all the data points about the contract into a table. This table may include columns such as Contract Owner, Coverage, and Confidentiality. The problem is that if I change any of these columns or add new columns later, I have to recreate the entire table to make sure all the entries are in the correct format. am. Also, some contracts require specific columns, while others do not. But filling every column for every contract is a huge waste of time.
On the other hand, an ontology about contracts might have classes such as “business contract” and “tenant contract”, each with its own properties.think more like Tree diagram than a hard table. If you want to add a new property, it’s as easy as adding a branch in the right place. You might want to add a branch called Student Lease Agreement under Lease Agreement. However, something like a business NDA has absolutely nothing to do with student rental agreements, so it doesn’t make sense to add this as an entire column to all types of agreements.
Ontology is also extremely Useful for machine learning. It’s hard even if it’s big language model To understand all this: that “Paris” is a city and therefore has certain characteristics, that you are not in this city, but you might want to be in it, so , that Paris should suggest a suitable route to Paris you. If you’re using an ontology, all this information feeds directly into your machine learning model. In this way, the model can focus its functionality on suggesting optimal rail routes and tourist destinations.
Ontology modeling and the Semantic Web
Ontologies are semantic webIt’s basically a fancy word to express that we want the web to be human readable and work with linked data, rather than a scattered bunch of https URLs pointing to each other. am. For example, searching for ‘Paris’ will not only return a list of links to pages with the word ‘Paris’ in it, but also give you pertinent information about the city, its inhabitants, and how to get there. there you go
What is the Semantic Web?
The Semantic Web reflects the idea that the web should work with human-readable, linked data, rather than a scattered https URLs pointing to each other. Data ontologies are a key part of enabling this new and improved web.
In the Semantic Web, distributed and heterogeneous databases can speak a common language and interact with each other. Distributed in this context means that the database resides on many different servers. Heterogeneous means that they may differ in terms of architecture, data format, etc. But what these databases have in common is that they know everything about what people and cities are, for example. So a database about the world’s largest cities can be matched against a database of the world’s richest people without having to deal with many technical problems. And when you run a query, you can quickly see which major cities have the richest people.
In more sophisticated technical terms, ontologies and the Semantic Web ensure interoperability, cross-database search, and smooth knowledge management. Interoperability means that databases can work with each other. Cross-database searching means that you can search results in multiple databases at once and infer logical conclusions from them.smooth at the end IP management This means that information is stored and used in a simple and user-friendly way. Think about how search results and social media feeds have changed over the last few years. The Semantic Web is well on its way.
Practical application of ontology
We’ve talked a little bit about web search, but ontologies are much more than that.In the pharmaceutical industry, AstraZeneca uses ontologies to test initial hypothesesThey build large datasets according to the principles of ontology (i.e. there are different things like proteins, genes, diseases etc. and they have specific relationships to each other). This data set came with a user interface that allowed AstraZeneca researchers to explore everything and their relationships before starting drug development.
Another use case helped people with medical records being ontologically organized. make better food choicesIn yet another case, using financial data exposing financial crimes.
What all these applications have in common is that they are user-centric and based on real-world problems. The Internet is becoming less concerned with “Hey, why is this URL broken?” “Hello Internet, I have a problem. Can you help me?”
Age of ontology
Ontologies may be sold as: next big thing of data scienceIn fact, it’s an age-old discipline. The idea of using it for data is indeed disruptive. However, wherever you look, it’s already expanded to data.
The question you are asking now should no longer be “What is an ontology and why do we need it?” Rather, “Why isn’t my company working with ontologies yet?”
You don’t have to pick up a philosophy book to answer that question. However, we encourage you to take a critical look at how your company is currently doing this and reflect on how an ontology can improve your current business processes.
You should not jump on all buzzwords and technology fads (learn NFTs?). But we must respect the age-old principles that revolutionize our technology.
