by Gilles Grousset, CTO at Backelite
A year usually starts in January. For us, it starts in May with the Google I/O…
Machine Learning : one step beyond
Machine Learning is an application of Artificial Intelligence (AI) that enables computer software to learn and process results without being explicitly programmed to do so.
Although Machine Learning is not new, thanks to increasing computer power anddata collection, its applications are more and more relevant: from picture recognition to traffic predictions, it is now used to solve several complex problems.
Thanks to advanced tooling and cloud platforms continuous improvement, it is easier and easier to add Machine Learning features into applications without the need of an in-depth knowledge of Machine Learning algorithm.
So there seems to be no reason why Machine Learning will not be THE tech trend of 2018 !
Some use cases…
Services and apps that can learn from your behaviour, and adapt to the context / usage.
Example : personalized movies & TV series suggestions on Netflix.
Learning from large quantities of collected data to help making critical decisions.
Example : Google DeepMind Health in UK helps lower the rate of patients dying from curable conditions.
Better than humans?
Machine Learning algorithms are purpose-driven and can perform better than humans when it comes to design.
Example: in industrial design, Machine Learning can come up with stronger parts using less material than a human can do.
Chatbots are so 2017 : in 2018, think Vocal Assistants
2017 was definitely the year for chatbots, and many single-purpose bots were deployed.
In 2018, chatbots evolve to vocal assistants.
A vocal assistant is not only a “talking chatbot”, it is multi-purposed (has several skills), is available on several devices (smart speaker, smartphone) and will soon interact with other humans for you.
Apple came last in the business of smart speakers for homes with its HomePod, but now the competition is in full blast between Apple, Amazon (the pioneer with Echo) and Google (with Google Home). With new use cases everyday, the adoption of those devices should increase massively in 2018.
Some use cases…
A seamless service integration
Over the years, manufacturers have been opening their assistant to third-party applications and services.
Example : smart home devices control (many devices supported), pizza ordering with Domino’s, ride reservation on Uber…
Your assistant can do it for you
Thanks to NLU’s (Natural Language Understanding) and NLG’s (Natural Language Generation) fast improving, vocal assistants are getting better and better at interacting with real humans.
Example : during last Google I/O, the impressive demo of Google Duplex, where a vocal assistant was able to call and take a hairdresser’s appointment all by itself.
Blockchain is here to stay!
Blockchain is the technology behind the cryptocurrencies hype.
But above all, it is a very promising technology.
In a few words, the blockchain technology can transmit data in a secure way without the need of a trusted third-party (like a bank for example). It is the beauty of decentralization.
Some may see it as a threat (businesses based on a centralized system), others as an opportunity to create game changing services, but it is here to stay for a longtime!
Some use cases…
“Code as law”
Smart contracts are tiny programs executed on the blockchain, and are intended to digitally facilitate the execution of a contract.
How? By removing third parties and contractual clauses.
Example : Fizzy is a flight insurance proposed by AXA running with smart contracts. If your flight is delayed by more than 2 hours, you automatically get your compensation without the need of filling any paperwork.
A better traceability
Pharmaceutical and food industries need to carefully trace their supply chain.
Using blockchain for this task can make traceability easier and safer.
Example : the French retail company Carrefour has started using blockchain to trace his chicken supply chain.