Our unsupervised learning model

Capturing, refining and providing valuable output through fashion digital interfaces

Background: Chicisimo built the interfaces and infrastructure to help people find what outfit to wear, and what garment to buy. We focused on (i) creating input interfaces that allow people to express a need; on (ii) building the data system that understands and responds to that need, and; on (iii) creating an environment where users are incentivized to provide data, specially descriptors of outfits and garments, to keep the learning loop going. 

Process: We approached the above opportunity by building a small team of product people and engineers with a common understanding of the what-to-wear problem, seeking to build interfaces with the following characteristics:

  1. The interfaces we build must allow people to express a what-to-wear need, from objective (type of garment, color, fabric, print…) and subjective (styles, occasions, age…) points of view; 
  2. The development of interfaces and data infrastructure influence each other;
  3. For 1 and 2 to happen, we have to uncover the organic data structures that arise from said interface;
  4. While users provide input and respond to output, the entire experience encourages them to provide further input. Input comes in the form of (i) expressed taste, such as expressing what type of style they like, expressing what occasions are relevant to them in terms of clothing, what type of clothing they like or own. Input also comes in the form of (ii) categorizing content, such as grouping outfits into described albums, tagging outfits, establishing correlations among items, providing relevant queries, etc. This input supplies our team with feedback to build more effective interfaces and accumulates data so we can respond to more complex inputs with it;
  5. The data structures are refined through unsupervised learning and collaborative filtering, by allowing us to find new data structures and ways to aggregate the data which aids in the consequent interface decisions. A learning loop is created.

As a result of the above, our product experience leads people to  create content for us, search for it and interact with it. It is this community who, by creating content, categorizing it, and searching for it, allows improving our understanding of the data and creating new algorithms and experiences for the community. 

Automatic curation. Our system encourages content creators differently based on the content they create: As not all users are as good as tagging or describing their content, we’ve defined an approach to easily classify users based on their quality content, and we give a different weight to the information provided by different users. Some users are featured in prominent places through our consumer apps, creating a positive feedback loop where they get rewarded for posting more and better content. We also monitor outfits with high engagement and add them to this list of outfits if they meet certain criteria.

Future iterations: There are several ways in which our process can be improved, and we have other sources of data we can use to improve or create new experiences:

  • We are in the early days of allowing people to describe/categorize the clothes in their closets. We are in the early days because we have not needed to be at a later stage. This will change;
  • We are not using the style of outfits people like, when returning results to their queries; 
  • We are not using the closet data to help people shop for new products. From the point of view of our data structure, a closet is equal to a catalogue, so obtaining shoppable clothes that correlate with items in a user closet something we do easily; 
  • We also know we can have the Graph attach further descriptors to certain outfits. We’ll expand the collaborative filtering by adding the descriptors users are querying for when opening an outfit or product. For example, if an outfit shows up in a query including “red pants”, and is constantly being opened, there is a high probability that the look is about “red pants” too. Albums names could provide additional descriptors to outfits, and it has not been a priority for us; 
  • We are not using contextual data to uncover new possibilities, or to influence the output, like geolocation, time of the year, weather.

Clothes classifier

We have built a system to “classify clothes” automatically, with 175 million classified and correlated meta-products. This system allows us to automatically understand, manage and act upon any collection from any retailer (similarities, correlations, recommendations…)

To explain it easily, this system is like a brain that understands clothes and outfits, and allows you to organize and display products at your convenience, or your shopper’s convenience. It was constructed after analyzing millions of perfectly described outfits and fashion products uploaded to our system by different subsets of users of Chicisimo, and after analyzing how people interact with them.

The system converts fashion products into meta-products, which are abstractions of specific products of any catalog or closet. A fashion product is ephemeral, but its descriptors are not, so the system retains the value.

A meta-product is the most basic yet relevant description of a product, and one of the first tasks of our infrastructure is to convert any incoming fashion product into a meta-product. While a person might see a given garment, our system reads a set of descriptors, for example: burgundy + sweater + v-neck + comfy + casual + for school + size 42 + cashmere, etc.

For any given retailer, this system can automatically digest its catalogue and then, automatically: (i) Understand each product; (ii) Identify missing information; (iii) Identify similar products, defining similarity in a number of ways; (iv) Build complete looks mixing and matching the clothes in the collection; (v) Identify products that make sense to display together; (vi) Recreate any outfit with garments of the catalogue; (vii) display the correct products for each shopper, or for the current interest of each shopper; (viii) If the system detects a product that it cannot understand, it isolates the descriptor and incorporates it into the fashion ontology if the team so wishes.



If you’d like to contact us, please get in touch with co-founder Gabriel Aldamiz-echevarría at Gabriel.aldamiz at


5 omnichannel examples of customers interactions

1. In-Bedroom Fashion Stylish. You will WOW your shoppers!

Read more here

2. In-Store Outfit Recommender. 2nd of the 4 Best Examples of Omnichannel eCommerce Personalization in Fashion Retail

Read more here.

3. Smart Fitting Room with Clothes and Outfit recommender

Read more here.

4. Digital closets

Read more here.

5. Adding your clothes to your digital closet

This approach works together with other approaches, not isolated. Read more here.


Our Manifesto

Our Manifesto for the Future of Fashion Retail

Technology is opening up a lots of options for fashion and we want to be very clear about what we consider the true building blocks are. 

Here’s our view, our Manifesto, for the Future of Fashion Retail:


Each fashion retailer will own its Taste Graph like Pinterest does.

This Taste Graph will contain the intelligence generated by shoppers and their interactions with products. A Taste Graph is your unique data asset and engine to build a competitive moat around your business.

Read about the Taste Graph.


Retailers will build Taste Profiles of each shopper like Spotify or Netflix.

A Taste Profile summarizes the taste of an individual shopper, what clothes she has in her closet, and what are the drivers behind her purchases.

Read about Taste Profiles of shoppers.


Ontologies will include all fashion concepts, even non-physical clothes descriptors but very relevant when deciding an outfit.

Each retailer will manage its catalogue and its shoppers with their own ontology.

Read about the new ontologies and what they allow.


Seemingly unimportant services that look like toys will win the heart of people, their data and their pockets.

These toys might look like In-Bedroom Fashion Stylists (watch video), In-Store Recommenders (watch video) or Digital Closets (watch video and read about the tech).

NOTE: We are not claiming that all fashion retailers are set for the above infrastructure. But it does apply to 100% of large fashion retailers.


Our blog posts

Thoughts on the future of fashion interfaces, data systems, and fashion technology

July 17th, 2019


May 21st, 2019


April 25th, 2019


April 4th, 2019


December 29th, 2018


January 30th, 2018



Compliance at the Fashion Taste API

We at the Fashion Taste API respect the data concerns of our customers and have committed to making sure they can use our taste profile and personalization technologies, safely and compliantly. We are committed to the confidentiality, privacy and security of our customers data.

The Fashion Taste API works with respected teams

At the Fashion Taste API, our team believes that the ultimate customer experience relies strongly on transparency and integrity. The Fashion Taste API works with respected teams of IP and data privacy lawyers, financial experts and Venture Capitalists as venture partners.

We are dedicated to meeting global requirements and regulations when it comes to data protection, privacy, security, and operations. Having your trust as a partner in taste profiles and personalization is of the utmost importance to us, and we will continue to invest extensive resources toward maintaining the highest possible industry standards so you can use our technology safely, efficiently, and compliantly.

Fashion Taste API & GDPR

Fashion Taste API is committed to the confidentiality, data privacy and security of its customers and their end-users. We are and will continue to invest extensive resources towards maintaining the highest levels of data protection, privacy and security standards. We are compliant with applicable laws and regulations, and are committed to our ongoing compliance with the EU General Data Protection Regulation (GDPR) and related guidelines.

  • Data Storage. Fashion Taste API stores customer data in Ireland. Fashion Taste API has proxies deployed in Sao Paulo and Oregon to reduce latency and increase response time in those areas.

  • Satisfy the Definition of a Data Processor- “Processes Personal Data on behalf of the Controller”. Fashion Taste API provides a full fledged platform which allows marketers to segment and target users with personalized offerings across web, mobile, email and other channels. As such, Fashion Taste API processes its customers’ end-user data, thus, Fashion Taste API customers who are deemed as “Data Controllers” under GDPR and the end users would constitute the “Data Subjects” whose rights must be protected.

  • Act on Behalf of the Controller Based on Controller Authorization. Fashion Taste API only collects information based on a duly executed contract from the controller. Once an agreement to process data is terminated, Fashion Taste API ceases collecting personal data from the customer’s sources, and the records are deleted within a reasonable period of time from when a deletion request is made. The purposes of the data processing are determined by our customers, i.e. the Controllers.

  • Appointment of a Data Privacy Officer (DPO). Fashion Taste API has appointed a Data Privacy Officer who oversees our privacy compliance and development program.

  • Integrity and Confidentiality. Fashion Taste API has employs appropriate technical and organizational measures to safeguard Personal Data.

  • Demonstrate Compliance with GDPR. Fashion Taste API keeps a record of its processing activities carried out on behalf of the controller, its DPO is open for questions and data processing addendum is entered into with every customer.

  • Lawfulness, Fairness and Transparency. Fashion Taste API collects and processes Personal Data lawfully and is transparent with its customers about its processing activities. Fashion Taste API has entered with its customers in order to maintain the legal basis for the processing (usually “consent” or “legitimate interest”), which is warranted by the customer as the Controller.

  • Processing Personal Data- Purpose Limitation. Fashion Taste API only collects data which may be used to analyze user behavior and to provide personalized experiences. Fashion Taste API does not combine any customer collected data with data collected from other customers, does not determine the purpose of processing, and does not share data with third parties except where required to by law.

  • Processing Personal Data- Data Minimization and Proportionality. Fashion Taste API does not onboard any data which is unnecessary or disproportionate to its needs to best serve end users with personalized experiences. CRM and other data may be onboarded and Fashion Taste API expects its customers to only onboard data which satisfies the proportionality and lawfulness requirement. At the point when Fashion Taste API provides a mechanism for onboarding customer data, we request that sensitive or payment data not be onboarded without our consent.

  • Processing Personal Data- Accuracy. At the point when Fashion Taste API provides a mechanism for onboarding customer data, Fashion Taste API will allow its customers to rectify any errors or misapplications in onboarded data with new CRM data pushed by the controller.

  • Processing Personal Data- Storage limitation. Fashion Taste API does not store any data unnecessarily and (at the point when Fashion Taste API provides a mechanism for onboarding customer data) expects its controllers to refrain from onboarding data which is unnecessary for personalization purposes

  • Processing Personal Data- Accountability. Fashion Taste API’s DPO will work to continuously optimize and introduce improvements and customer feedback to the Fashion Taste API privacy program. The DPO will also cooperate with controllers in case of inquiries and data breaches.

  • Data Subjects Rights. Erase, Rectify and Export Personal Data. Fashion Taste API will cooperate in full with controllers for handling requests regarding their end users’ data.

  • Information Provisions- Transparency. Fashion Taste API will cooperate with controllers regarding data requests from customers.

  • Ownership of Data, Process and Tool. It is now gone the time to provide personalized recommendations to people, without understanding their taste. The Fashion Taste API has been designed to empower teams to own the process, the tool, and the data, while maintaining the highest levels of data protection, privacy and security standards.
    • DATA PRIVACY: The Fashion Taste API is committed to the confidentiality, data privacy and security of its customers and their end-users. We are and will continue to invest extensive resources towards maintaining the highest levels of data protection, privacy and security standards.
    • PROCESS & TOOL: The Fashion Taste API does not interfere with any of your existing processes. It starts registering data, as easy as installing Google Analytics. You speed up with whenever you want, with the advantage of having already learnt from the captured data. The fashion industry no longer believes in black-box approaches to personalization that provide zero ownership and control to teams. The Fashion Taste API is a tool for your teams to use on growth. You control the tool.

Vertical Machine Learning

A Vertical Machine Learning approach provides a competitive moat around your business


We at the Fashion Taste API follow a Vertical Machine Learning approach to fashion. We believe in the unique advantage of having a profound understanding of the consumer problems, end to end.

With a focus on the consumers, we’ve applied machine learning to automate the two processes that we’ve considered game-changing:

  • Capturing shopper’s taste data (including what clothes they have in their closet);
  • Building the data infrastructure that allows us to understand input automatically, and to deliver personalized output also automatically.

Read our Manifesto on the Future of Fashion Retail.


We believe that fashion retailers have a clear advantage over newcomers: they are the domain experts. The specific benefits, however, need to be captured:

  • Fashion retailers have a profound understanding of the fashion consumer;
  • Shoppers generate a unique body of data around the retailers properties. Retailers can choose to start capturing that intelligence and build their moat;
  • Physical stores are a unique layer of experimentation to test omnichannel solutions and receive fast feedback.


At some point, newcomers will offer In-Bedroom Fashion Stylist or Smart Mirrors to end consumers.

Apart from the increase in market share that this new category will generate, In-Bedroom Fashion Stylists will have two very relevant consequences:

  • Data: The amount of incoming taste data will grow exponentially;
  • Learning: The speed of learning will accelerate thanks to better shopper feedback.

And the current competitive moat around traditional fashion retailers will be gone.


The more digitally powered experiences you offer, the more taste data you will capture. The more data you capture, the better you can automatically serve your shoppers. The better you serve your shoppers, the more adrenaline you generate in your teams, the more ambitious they become.


Foundational patents in capturing and understanding fashion taste

Patents expected to provide a competitive advantage


We’ve been analyzing fashion taste for more than a decade.

As a result, we own foundational patents in the online fashion market. An independent review of the Fashion Taste API portfolio uncovered relevant market adoption related to some of our patents.

This intellectual property powers our Digital Closet Technology, our In-Store Outfit Recommender, our In-Bedroom Fashion Stylist and other solutions.


We are all starting to see fashion images with dots on top of them, these dots enabling you to find and buy the products in the picture. This system is gaining popularity since 2016, and it was patented by us in 2013.


In our prior work with music, we extracted relations among songs by analyzing the sequence in which people would listen to music, and by analyzing our interpretation of items. Apple later bought the systems and patents for Apple Music. We then built similar systems for personal finance, and for content-agnostic input.

With this background, we thought that an outfit was like a playlist of songs: a way to communicate that it makes sense to consume several items together. This simple concept allowed us to conceptualize the first version of our Social Fashion Graph. In 2013, we protected a system to capture and understand how people mix and match clothes in outfits and closets.

Today, the technology has evolved significantly. It captures any input in our our Ontology, which includes any concept relevant to fashion, and assigns descriptors to products and to people. We can extract all type of relations among analyzed items.


Our initial interest was related to social tagging of images. Then, we invented more complex mechanisms to assign descriptors to clothes, outfits and people. Our IP protects a system to search outfit ideas and fashion images.

It could be a simple query such as “give me ideas to wear jeans” or a more complex one such as “give me ideas to wear my pair of red pants with my floral shirt, to go to the office in winter”, or “give me clothes that are “similar” grouped by your concepts of similarity”. This is strongly related to our Fashion Taste Graph Technology.

The Fashion Taste API patents are expected to provide a competitive advantage.


Our fashion ontology

Our fashion ontology is our tool to understand products and people, and we built it to respond to very specific needs in our mobile apps.

Fashion ontology
Multilevel fashion ontology to describe clothes and user needs

Why we built our fashion ontology?

Fashion lacks a standard to classify clothes or to refer to the variety of concepts that describe products, styles, and personal fashion preferences.

At some point, we found ourselves receiving a lot of data via our consumer apps. We were receiving millions of described outfits, the described clothes in millions of closets, and hundreds of millions of what-to-wear queries from people trying to decide what to wear

Looking at this data, we only saw unorganized data, so chaotic that it was impossible to manage or build on top of it.

Our users could not really express their needs precisely in a ways that we could understand them, we couldn’t even describe the content displayed in our apps, in a way that could be found by those in need of it.

Our iOS app, in action

The backbone of our Taste Graph

Today, our fashion ontology is the backbone of our Taste Graph technology, and it is core to our development process.

We divide our ontology into two parts:

  • Products ontology. It is a 5-level ontology that describes products and subjective characteristics of products;
  • Outfits ontology. It is a 2-level ontology that describes outfits, mostly with subjective descriptors.
Our Alexa app 🙂

Our fashion ontology today

Our fashion ontology is our tool to understand products and people: it goes well beyond assigning descriptors to clothing. It has a very strong emphasis on describing people.

Our ontology today is the classification of descriptors needed to define a fashion product and a shopper’s need, in terms that are relevant for shoppers, for retailers and for data scientists. It contains the vocabulary used by people and by fashion retailers, with an understanding of how these descriptors are used together. Our ontology covers all aspects related to fashion purchase drivers: product properties, occasions, brands or influencers, or trends: It is formed by +2500 unique concepts and a few hundreds of thousands of descriptors.

Lots of interesting processes in this video – all related to our ontology

What we’ve built on top of the ontology?

All our Chicisimo products have been built on top of our Fashion Taste Graph, and the ontology is it’s backbone. This is what the ontology has allowed us to build:

  • Smart Virtual Closet Technology
  • Taste Graph
  • Search tech
  • Recommendations is all forms and personalization
  • Clothes classifier
  • The ontology has also provided a unique common language for the team to communicate, and work, more efficiently.
Another one of our processes

How can I fashion ontology help your business?

Well, that’s a question you’ll have to answer prior to building one. You might already have a fashion taxonomy, and you’ll expand it little by little as your product team demands more tech strength.

In general, your fashion ontology will help you describe products and shoppers in a way that allows you to:

  • Understand products and shoppers;
  • Manage products and shoppers.

Your ontology will help you manage your catalogue with new browsing methods, humanized categories adapted to different types of shoppers or situations. It will reflect how your organization wants to talk to shoppers, always retaining total control over your voice while understanding the voice of your shoppers.

It will guide your data-attaching mechanisms to attach the right descriptors to products and shoppers. Data-attaching mechanisms include your taste graph, your editors, your deep learning algorithms, and other sources of product and shopper descriptors.

Characteristics of our ontology

Again, it will depend on the state you are at. Your ontology or fashion taxonomy is a tool that contains the organized and structured descriptors that define your products and your shoppers, in the terms that are relevant for your organization and your shoppers.

  • Your ontology includes clothing and non-clothing descriptors;
  • The ontology includes descriptors used by your organization, and by your shoppers;
  • Descriptors are classified in categories, types, properties and more;
  • We capture the correlations among your descriptors as defined by your products and your shoppers.

How can you use your ontology?

The Fashion Taste API main focus has been about the search for structure, in the unstructured world of clothing classification.

Once we learnt about clothing classification, we learnt that purchase drivers go well beyond clothing taste, and we gave a structure to those as well. So, believe us, we love clean and structured data that our algorithms can work with.

Despite of the above (or because of the above), we have found amazing value in shoppers’ unstructured input. This input are descriptors that you would not expect your shoppers to be using when interacting with you.

We’ve learnt that people refer to their fashion needs in ways that are very different to how retailers do. Also, the type of taxonomies that fashion retailers tend to have cover mostly product descriptions. But drivers of purchases are related to other factors: products, occasions, influencers, brands and trends. You can read about it here.

Need help?

Contact us