Primal is a semantic engine

Technology

Primal is the first computational semantic engine that synthesizes data about individual interests. This disruptive innovation is the product of over six years of R&D.

We call our innovation semantic synthesis. It is a generative approach to knowledge representation that manufactures rich semantic data in real-time, requiring only sparse data sources as inputs.

Primal is backed by a growing portfolio of intellectual property.

For frequently asked questions about our Technology, please see our FAQ.

The Problem: Big Data personalization is expensive & complex

Personalization requires data that describes the interests of individual consumers. This data is difficult and expensive to provision.

Conventional solutions build this data manually or derive it indirectly using large scale statistical analyses of existing content or user activity (big data).

These conventional approaches are much too costly for ubiquitous personalization. Manual methods are prohibitively expensive at scale, so large projects necessarily rely on computing machinery. However, few companies possess the infrastructure or big data needed to represent the full breadth of individual interests.


Figure 1: Primal disrupts a fundamental barrier to personalization

The Solution: Semantic Synthesis

Modeling knowledge generation, not modeling knowledge

Our insight into the solution is deceptively simple. Where data-driven approaches impose a barrier, Primal works effectively with very small data. Since data analyses impose a barrier, Primal leverages data synthesis. And where hybrid solutions are prohibitively complex, Primal is simple enough for any developer to use.

The core differentiating aspect of our technology is its ability to synthesize rich semantic data on-the-fly, even if you don’t have much existing data to leverage.

Primal’s computational semantic engine synthesizes rich semantic data about individual interests in real-time.  The semantic data can then be used as machine-readable inputs to a series of automated processes and software agents to get things done faster, easier, and simpler than with today’s complex and costly solutions – all without having to change a company’s current infrastructure.

Much like human beings use words in combination with grammatical rules to form statements, semantic synthesis uses a vocabulary of atomic semantic data and a proprietary set of generative rules to synthesize semantic networks, as shown in Figure 2. Primal creates these machine-readable interest networks on-demand, requiring only simple indicators of user interests.


Figure 2: Primal models the process of knowledge generation

Semantic synthesis bridges the gap between the sparse expressions provided by consumers and the rich, structured data needed to direct the activities of computers. In this way, Primal extends existing computer systems to perform highly personalized tasks for individual users.

Primal’s Advantages for Rapid Product Innovation

Primal is the only personalization infrastructure capable of powering rapid product innovation for third-party development. Features include:

  • A simple and intuitive interface that makes it easy for any developer to use this sophisticated computational engine, even if they have no prior experience with knowledge engineering.
  • A real-time, computational approach that avoids the monumental costs of big data analysis that has frustrated efforts in the past.
  • Works with unstructured and semi-structured content on the Internet as if the Web were already semantic.
  • Primal’s computational engine can be used without changing your existing search or content management infrastructure.
  • The interests of consumers may be expressed directly (as in a search query or a social profile) or contextually (as in browsing behavior or sensing data from a mobile device).
  • The interest networks may be persisted as a valuable asset to you and your consumers.
  • Primal learns more about the interests of consumers over time, so you do not have to continually maintain the applications that are Powered by Primal.
  • Primal’s Assistants (software agents and distributable components) and content extensibility power a broad range of sophisticated knowledge engineering solutions.

How It Works

Human beings do not retrieve their knowledge from knowledge bases, like conventional data-driven computing systems. Instead, we generate (synthesize) our statements on the fly. We use a relatively small vocabulary of words (small data) in combination with grammatical rules (computational rules) to form more complex expressions.

Similarly, Primal has developed a computational approach to synthesize more complex semantic data from a much smaller “vocabulary” of proprietary data. Primal’s technology uses its proprietary “grammar” of generative rules to synthesize complex semantic data (“expressions”) in real-time.

Just as human vocabularies are a mix of common and specialized language, Primal’s vocabularies can be easily and inexpensively expanded to include specialized areas of knowledge such as health, finance, and politics.

Just as humans can express themselves in limitless ways, Primal’s computational approach is what allows it to generate data for a limitless range of knowledge domains and applications. Primal creates this expressive, machine-readable semantic data on-demand, using the same simple “language” used on the Web for addressing documents and services.


Figure 3: Primal turns simple indicators of interest into expressive machine-readable data

Primal’s semantic engine has both offline and real-time components: an analysis engine and a synthesis engine (see figure 3).

Unlike conventional approaches, Primal’s process starts with the end-user. It takes simple indicators of interests and uses this context to synthesize data structures that represent these interests in a way that machines can process. These user inputs are simple and intuitive word associations, congruent with the way people make meaning.

The interests of end-users may be expressed explicitly (as in a search query or a social profile) or implicitly (as in their browsing behavior or sensing data from a mobile device).

The output of synthesis is a semantic representation of the interests of end-users. These interests can now be processed by machines and software agents to conduct tasks on behalf of the users, much like personal assistants.

The possibilities for these software agents are endless: searching for information, making social connections, creating reports, tracking topics of interest, across whatever subjects end-users may express.

These semantic representations of interests can be disposed of immediately after the task has been completed, or they can be persisted as a knowledge-base for each individual end-user. This data can also be fed back into Primal’s semantic engine, providing a learning mechanism. With each activity and task, the system becomes more attuned to their interests, just as a personal assistant becomes more helpful over time.

The analysis engine drives the offline activities. This engine analyzes representative content and creates vocabularies of atomic semantic data, like the words in a dictionary. These vocabularies are used as the building blocks for more complex semantic representations in the real-time synthesis operations.

Just as human vocabularies are a mix of common and specialized language, Primal’s vocabularies can support specialized vocabularies that cover large areas of human knowledge like health, finance, and politics.

The real-time synthesis engine is where the vocabularies of atomic semantics are assembled into complex semantic representations of meaning.


Figure 4: Primal’s real-time semantic engine

Conventional semantic technologies are challenged by the Internet’s scale and pace of information production.

Much of the effort in semantic representation has been focused on annotating existing content. Creating this type of semantic layer over existing content is proving to be a daunting task due to the sheer glut of online content and the compounding effect in the volume of data needed to create machine-readable semantics.

Figure 4 illustrates the real-time aspects of Primal’s semantic engine that contrast with conventional approaches. Note that the synthesis of semantic data is largely decoupled from the retrieval and annotation of content from the Web, fundamentally changing the cost-performance structure of the solution.

Primal’s synthesized semantic representations of user intent allow us to interact with unstructured and semi-structured content on the Internet as if the Web was already semantic. Text analysis is used to categorize and semantically annotate online content within the organizing frame of the individual user models created by Primal.

The major difference here is that the content organization is discovered through the expectations of consumers, rather than being imposed by knowledge engineers in advance. In other words, Primal’s semantic webs evolve as a by-product of this consumer-directed process, avoiding the bottleneck of semantic annotation that has frustrated efforts in the past.

Technical Components

Primal’s technology platform features a number of key components, illustrated in figure 5:

  • The Analysis Engine (1) maintains a common vocabulary of atomic semantic data that allows for interoperability across different knowledge domains. It also maintains specialized vocabularies that cover large areas of human knowledge like Health, Finance, and Politics.
  • The Synthesis Engine (2) uses a proprietary set of Knowledge Generation Rules (3) applied to the atomic semantics created by the Analysis Engine to manufacture new, explicit semantic data on demand.
  • The Actions Framework (4) provides extensibility for mapping content sources, including demanding high-volume and real-time sources.
  • Software Agents (5) use the semantic data created by Synthesis with the content provided by the Actions Framework to perform and automate tasks on the open Web.
  • Semantic User Models (6) store semantic data that models the interests of individual users, which can be used to further personalize the data generated by the Synthesis Engine.
  • Complex Adaptive Feedback (7) provides a mechanism for learning interests over time, making the system more aware of the community of interests and automating the maintenance of these interest networks.

Figure 5: Technical components in Primal’s semantic engine

Intellectual Property

Primal has a rapidly growing portfolio that includes 6 US patents granted, 22 US pending and 30 foreign pending in the areas of knowledge representation, search, social networking, advertising, and media.

Primal has secured a blanket of protection through a patent pool tailored to safeguard key components of its platform architecture, as well as platform-enabled solutions such as user modeling, semantic search, advertising, and content integration.

Here is a small sample of Primal’s intellectual assets: