Leapsight Semantic Dataspace
LSD is a distributed, NoSQL, relational and deductive database that was purpose-build to provide the foundation for scalable, interactive and intelligent knowledge-based cognitive systems.
Benefits
PLasticity & Agility
Extreme data and business logic modifiability and reduced time-to-market, without sacrificing expressiveness.
LSD enables the rapid evolution of your knowledge base and the ingestion of data from new data sources making these changes immediately available for operational and analytical use.
LSD embeds Leaplog, our logic query and programming language, which gives you the capability to change business logic in minutes to adapt to new business contexts in record time.
Real-time Actionable Knowledge
Knowledge is actionable only if you have the capability to act upon it when and where required.
LSD enables a shared knowledge backplane between cognitive (learning) and automation (exceution) processes by using knowledge graph technologies.
Dynamic logic rules are used to deduce new knowledge, allowing the implementation of the real-time intelligence required for personalisation, mass-customisation and other cognitive use cases.
Web Scale
LSD provides low-cost scalability through a masterless, share-nothing, highly concurrent architecture.
LSD can scale easily and without downtime. Simply add more nodes to the cluster to increase reliability and performance.
Use Cases
Omni-touchpoint brand engagement
Consumer Brands, Retail and Services increasingly need to establish personal relationships with 100s of millions of consumers.
LSD’s architecture is ready for the scalability and continuous availability required by consumer engagement.
It acts as a shared knowledge space between all consumer touch points, providing not only the aggregation of multiple consumer profile data, but also the business logic to drive each consumer interaction consistently across touchpoints.
Avoid segregating your data across multiple technology providers such as CRM, Social Login and Personalisation services.
LSD knowledge graph capabilities and our Consumer Engagement solutions enable the consolidation of your consumers’ multiple profiles and all behavioural data to drive relevant engagement interactions.
Real-time Inbound Mass-customisation
As segmentation and outbound push effectiveness diminishes, brands need to start customising brand experiences in real-time.
LSD enables the customisation of brand experiences down to the individual consumer in real-time.
Its dynamic rule-based business logic provides the brain of your consumer engagement platform, combining data from multiple sources to generate a knowledge graph for each individual consumer that drives the next relevant actions and recommendations.
The set of rules to use in each consumer interaction can be defined at runtime, which allows the implementation of sophisticated multi-variate tests.
Real-time location-aware retail engagement
iBeacons are low-power, short-range transmitters that work on BLE (Bluetooth Low Energy) and enable micro-location and context based services for mobile applications. The iBeacon experience consists of the mobile application that talks to the beacon (hardware), the mobile operating system on which it is present, and the consumer engagement platform software which manages the beacons and feed the raw data, which then gets analysed and converted into specific customer notifications or campaigns.
Without an intelligent backend, iBeacon interactions can quickly become spam, but by allowing an LSD-based solution to drive consumer interactions based on consumer needs and preferences, you can ensure each individual consumer is engaged only in relevant and timely interactions.
Gamification – Consumer Engagement Mechanics
Engagement mechanics (game mechanics) are the core of an effective consumer engagement strategy. LSD was purpose-built to enable and glue together engagement mechanics such as rewards, loyalty programmes, raffles, instant-wins, and others. Moreover, by using new touchpoint enablers like iBeacons, an LSD-based solution can support innovative engagement forms of these mechanics, like in interactive treasure hunt where consumers need to visit certain locations (intra- or inter-store) to accumulate rewards.
Features
Knowledge Graph
LSD is a database of immutable, temporal and probabilistic facts implemented as a universal relation that uses the Semantic Web & Linked Data Standards as the basis for its information model.
In the real world, facts are immutable and occur at a particular moment in time. LSD was purposely build with this idea in mind, so once a fact is written in the dataspace, it cannot be changed. Updating a fact, thus, involves superseding the old fact by a new one.
The Web now provides us with multiple sources of information for an entity e.g. a consumer, so we need the capability to deal with ambiguous, conflicting and incomplete data.
That is why every fact also has a degree value, a number that denotes its veracity or probability.
Finally, a set of facts are stored as a knowledge graph (a labelled directed multi-graph), providing a natural and friction-less representation of many-to-many relationships, which are the nuts and bolts of consumer-brand relationships, social networks, product catalogues, IoT, recommendation systems and the basis for data isolation and provenance tracking required by Data Privacy regulations.
While very simple, this information model allows the representation of virtually any conceivable data entity freeing you from the limitations of the n-ary relations and their corresponding static physical schemas found in SQL databases.
Leaplog
As opposed to traditional deductive databases, rules can be bound at query time, providing a natural means of testing multiple business logic scenarios e.g. multi-variate testing.
To access and manipulate data, LSD provides Leaplog, a Datalog-based query and logic programming language. With Leaplog, business logic meets the plasticity and agility provided by the dataspace’s data modelling capabilities.
Leaplog enables the execution of dynamic business logic, deriving new facts about an entity based on custom defined rules. In addition, Leaplog can be used in combination with Semantic Web ontology languages to reason using semantics.
Fault Tolerance
LSD uses a masterless architecture and powerful self-healing capabilities to continue operating properly in the event of a failure of one or more of its components or outages caused by hardware failure or network partitions.
LSD protects your data by intelligently and automatically replicating data across the cluster. Even when a node fails or becomes unavailable to the rest of the cluster due to a network partition, data is temporarily replicated to a neighbouring node.
LSD will then periodically examine whether data resides on the correct physical node and hand them off to the proper node when possible.
Finally, Anti-entropy strategies avoids losing data due to node failure or network partition by actively comparing data amongst replicas and fixing any differences.
High Availability
Due to its data replication capabilities and masterless architecture, if a node experiences an outage, other nodes can continue to serve requests. In other words, the whole system remains available.
This “Always On” capability is key for knowledge-based applications, specially those driving Consumer Experience Customisation decisions in real-time.
LSD has been consciously designed to support this requirement and thus it prioritises high-availability of requests (read and writes) to replica (data) consistency. This is known as Brewer’s CAP theorem.
However, this is not an imposition. LSD offers the ability to tune the availability vs. latency trade-off in each request (read and write).
Scalability
LSD scales horizontally, automatically distributing data through out the cluster. Adding more nodes to the cluster not only increases capacity but also performance as LSD balances the computation of requests amongst the nodes in the cluster.
This coupled with the use of commodity hardware or cloud servers provides an inexpensive solution when compared with traditional SQL database hardware requirements.
Operational Simplicity
LSD is easy to setup and run. Operating LSD can be done without incurring a large operational burden and the tasks to be performed do not change while increasing the size of the cluster.
In addition, LSD provides self-healing capabilities that gives you piece of mind and reduces manual operations.
Get in touch!
We would love to learn about your needs. Enter your details and we'll get back to you shortly.