Archive for the ‘Use Cases’ Category

Bringing AI to its full potential with augmented intelligence

Monday, May 14th, 2018

It’s not even halfway through 2018 and the buzz around artificial intelligence (AI) is bigger than ever. It seems we’re finally starting to get past the overblown fear about a robot-dominated future, but the fear has been replaced by a lot of hype. AI is now being thrown around as a buzzword in every part of our lives.

If everything is AI, what exactly is AI anyway?

Like many others – including IBM – we think “augmented intelligence” is a better way to describe what we do with our AI platform. Primal’s augmented intelligence applications help people do their jobs better, serve individuals’ interests, and bring AI to its full potential. Unlike the fears of AI taking over everyone’s jobs, this kind of AI (augmented intelligence) is not a replacement for people – it enhances what people can do with their limited time and resources. The nail gun didn’t displace the carpenter, but carpenters sure are happy they have nail guns.

One of the use cases of Primal’s AI platform is an augmented intelligence application that can determine the meaning of content in order to find other content which is highly related. Primal does this much like a real human brain by using semantic synthesis (where semantics refers to the meaning of words, and to synthesize is to combining several things into a coherent whole).

Primal’s AI breaks textual objects such as an article, a post, a tweet, etc. down to an ‘atomic’ level. As an overly simplified example, “firetruck” becomes “fire” + “truck” and finds meaning by analyzing how this combination of words has been used across millions of online sources. It then synthesizes the meaning of each word (semantics) to determine the message being expressed in the original content.

In our firetruck example, the word “fire” could be related to wood, burning, and emergency, but also camping, coziness, and marshmallows. Only some of these words would be used in the same context as “truck”. Combining the words “fire” and “truck” together would net more relevant results such as burning, emergency, rescue, fire fighter, arson, etc. Again, this example used the simple term “firetruck”, but Primal has the ability to analyze and synthesize the meaning of a larger piece of content such as an entire news article.

Once Primal has analyzed the message of the content in this way, it can find and recommend other content that has the same meaning – not just similar keywords. All of this happens on-the-fly in real time, with each request to Primal creating a new set of potential connections and recommendations. Unlike big-data AI solutions which require lots of training data, Primal can find meaning from very small pieces of data, like a tweet.

Because it doesn’t rely on specialized databases, Primal’s technology can be used in a variety of different industries and use cases.

If you’re a social media manager, you know that curating and selecting relevant content to share on social channels is time-consuming, and it’s hard to break through the noise to get noticed by your target audience.

Primal’s augmented intelligence tools for social media can send individualized content to each person who follows you – not just random content, but highly relevant articles that fit in the overlap between their interests and those of your brand. Articles could be from third-party sources or from your own content library – enabling greater usage of created content.

When your followers receive an article just for them, it stimulates much greater interaction than when they read a generic post sent to your whole channel. Often it sparks a conversation, providing an opportunity to connect in new ways and build brand loyalty. They’re more likely to share the content further, which extends the reach of the original post and ultimately results in new organic followers for you. And all of this can happen automatically, freeing up your time and enabling you to connect to many more people.

Primal isn’t just for recommending articles or other content – it can also analyze product descriptions. When you run an online store, the product recommendations your customers see are usually based on what other customers bought. But it takes a lot of user data to teach the system to make meaningful recommendations.

If you’re not Amazon and you don’t have millions of customers, how do you get enough data to populate your recommendation database? With Primal’s augmented intelligence, you don’t need lots of data – Primal can analyze the original product description to determine meaning and recommend other highly relevant, complementary products.

These are just a few use cases where augmented intelligence can help you better engage with your customers, provide extra value, and increase your productivity. How could this work for you? Let us know by dropping a line to

Content Curation Website for Content Marketing

Monday, September 8th, 2014

A comprehensive resource for content marketing covering a wide range of topics of interest to content marketers.

Primal, a content discovery assistant, does most of the grunt work of finding the right information for the specific topics featured on the site.

Read on for a tour of the features of this content curation project. (more…)

Adding Automation to Human Content Curation

Thursday, May 8th, 2014

Content curation, the activity of collecting, organizing, and presenting information of interest to a target audience, is an essential component of content marketing.

As in many areas of marketing, aspects of content curation are being automated. Marketers are exploring how smart machines can help improve their productivity and performance. Automated content curation, however, is a young field and it’s difficult to separate the promise from the hype.

At Primal, we’re well aware of the importance of curation, but constraints of time, money, staffing left us unable to pursue sustained campaigns in content curation. As such, we wanted to explore how automated systems could be used to improve our efforts.

In this post, we’ll outline how we incorporated automated content curation into our activities and what we learned about its impact on marketing professionals. (more…)

User Modeling for Personalized Content Services

Tuesday, August 27th, 2013


Developers have long created software that customers use directly. But now, we’re creating solutions that incorporate internal representations of end-users and adapt to individual needs.

In this post, we’re going to introduce you to the most important component in your solution stack, the user model, explain why it’s so important, and show you how to incorporate user modeling into your solution.


Content Assistants for Messaging and Social Media

Wednesday, June 26th, 2013

Small conversational data such as tweets or text messages are a goldmine of individual interests. Millions of people everyday tweet about their favourite food, send a Kik or Facebook message about a recent TV episode, or take a photo on Instagram while attending a sporting event or concert.

However if you’re trying to recommend content or promote offers within a conversation, the ability to determine a user’s interests from these small data inputs is a huge challenge.

John Koestier in Venture Beat writes:

…tweets are difficult to register commercial intent…

If I tweet about my wife’s illness, are you going to target me with a random medicine? Or if you tweet about a great dinner you’re just about to eat, will you really be receptive to ads about a Greek restaurant just down the road?

In this post, we’ll show you an elegant solution for content recommendations, applied in messaging and social media platforms.

Personalized Web Content for Your Application

Wednesday, June 19th, 2013

Back in the early days of the Web, the Yahoo! Directory filled a huge gap in the ecosystem.

This portal helped us make sense of the Web. It organized content based on topics, connected those topics into a hierarchy to make sense of it all, and standardized the presentation of the content so that it was convenient and accessible.

But where can you find a portal that makes content from the Web accessible in your application? If you want to incorporate content from the Web into your app, let alone personalize it to the individual interests of your users, you’re undoubtedly confronting some daunting problems:

  • the content isn’t described in a way your application can understand;
  • the content isn’t organized in a way that matches your app; and
  • the content isn’t presented in a way that makes it easy to incorporate into your user experience.

In the sections that follow, we’re going to introduce you to a simple solution. Primal is like a personalized Web portal for your app. It’s a data service that makes it easy to incorporate Web content from sources you trust, tailored to the needs of each individual user of your application.


A Smart Machine as a Content Curator

Friday, May 31st, 2013

In this post, we’ll show you how to add Primal’s data service to your content curation solutions, as a fully automated, machine-editor.

We’ll also highlight the practical applications and benefits of using a smart machine as a complement to your manual and crowdsourcing strategies. Specifically:

1. How to filter out irrelevant content from your content supply.
2. How to provide personalized collections of content.

For our demo, we’ll use a content aggregator called Alltop, and show you how to recreate these examples and build a similar solution yourself.


Recommending Blog Posts within News Articles

Wednesday, May 22nd, 2013

In this post, we’re going to integrate Tumblr content into a Yahoo news feed, through the following steps:

Step 1: Create an interest graph about a Yahoo news article.

Step 2: Filter Tumblr content through the interest graph.

Step 3: Integrate filtered Tumblr content in the Yahoo news article.


Content Recommendations based on a Webpage

Wednesday, May 8th, 2013

How do you recommend products or content for a very specific interest at a moment in time?
In this post, we’ll show you how to build a world-class recommendation engine using Primal’s data service.

A Personalized News App

Thursday, March 7th, 2013

In the sections that follow, we’ll show you how Primal can support your product development efforts by providing the data that describes the interests of individuals and finding the content that people will truly love!


Social Media Analysis using your Interest Graph

Wednesday, September 5th, 2012

News alert! Your social network wants to get to know you. Unfortunately, identifying and understanding your interests based on 140 characters is pretty hard to do! That’s why marketers are busy trying everything to get to know you better.

In this post, we’ll demonstrate how Primal can be used to power social media analysis. This demos highlights tweets about topics you are interested in and helps you discover related topics that people are tweeting about right now!