Siri, Cortana, Google Now… Apple, Microsoft, Google… Virtual assistants are today a very promising development scenario for big technology. But not only big technology. There are also numerous companies with projects to universalize virtual assistants within the conversational platform sector, and what is known as the Internet of Things (IoT). One of them –possibly the best known– is Dialogflow, the application programming interface for the design of chatbots with natural language processing.
A large part of the income generated in the IoT sector will go towards developing applications, smart systems and devices, and platform-related services, including conversational platforms which set up a link between a mobile device, an application and a user interface. There are numerous studies on the future outlook of the Internet of Things, like the one prepared by the consulting company IDC. A forecast of the revenues::
According to IDC estimates, in 2013 there were 9,000 installed Internet of Things devices in the world. By 2020, this figure will have risen to 28 billion.
Although the IoT market is promising, the specific sector of virtual smart assistants is not far behind. A recent report by Transparency Market Research outlines some business forecasts for the period between 2014 and 2022: market value up from 580 billion dollars in 2014 to 5.1 billion in 2022, with an annual compound growth rate of 31.8%. According to the study, this increase is due to the spectacular boom in data and the fact that a number of e-commerce companies, banks and healthcare organizations need to improve their relationship with their customers through the use of chats and virtual assistants (chatbots) –although there are limitations in terms of costs and complexity.
Dialogflowi aims to revolutionize virtual assistants
Dialogflow enjoys enormous prestige within the virtual assistant sector, largely due to its great flexibility. Siri, Cortana and Google Now may be good technology products, each with their strong and weak points, but they can't do what Dialogflow does: enable third companies and their development teams to customize the design of their own assistants with a set of Software Development Kits (SDK) for Android and iOS operating systems.
These are some of the features its API REST can offer companies:
● Dialogflow is a platform that understands natural language. The service has basically three components: voice recognition (voice to text, text to voice, and automatic speech recognition or ASR through customized and dynamic language models), understanding natural language, and conversational capacity (the idea that the assistant understands the meaning and the intention of voice commands and generates appropriate responses through the different dialogue channels).
● Dialogflow has integrations for conversational platforms: thanks to this it can develop bots that are able to convert and distribute information in chat applications such as Facebook Messenger, Telegram, LINE and Kik, VoIP platforms like Skype, cloud services such as Twilio, and well-known productivity platforms like Slack.
● Dialogflow is a multilanguage solution: available in 14 languages.
The platform is based on a series of clearly defined concepts:
● Agents: a company's developers can train and try out an agent and then incorporate it in any application or device.
● Entities: the interpretation of user messages is based on key words or concepts that summarize the main ideas in each statement.
● Intentions: this is what allows the software to generate suitable responses to its interpretation of the natural language in the user's statements.
● Actions: the steps an application follows based on the user's inputs.
● Contexts: chains that represent the current context of the user's expressions. This is important for differentiating the exact meaning of a statement that may be confusing or have a similar meaning to others.
There are other companies such as x.ai, Clara Labs and Zoom.ai within the world of virtual assistants, focusing on organizing and scheduling meetings, supporting employees, collecting and distributing data on services like metrics tools such as Google Analytics or Mixpanel, on-demand CRM solutions like Salesforce, and productivity tools like Slack. This is the future of bots and assistants, and nothing they enable third companies to do could be done without APIs.
These articles will get you up to speed with the latest machine learning developments and advances and how they affect and could add value to the fintech industry and open banking:
In this article in Netguru, Timothy Clayton lists five areas in which artificial intelligence and, most specifically, machine learning are setting the tune for the development of new fintech products: customer service is redefined with bots and other AI interfaces; credit scores are faster and more reliable; trading and money management are enhanced; regulatory compliance is facilitated with automation; and the fight against bank fraud is made a lot easier.
In Medium, Ted Moses explains the most recent developments in deep learning applied to trading. By now, we have all heard about the robotization of financial trading globally; in this case, however, we go into a new level of sophistication. Deep learning performs extraordinarily in establishing patterns from data and later identifying these patterns from large amounts of data. Moses talks about the research done by Babak Hodjat toward what is already called "deep trading," a concept you must look out for. You will start hearing it more and more often in the immediate future.
This thread in Quora is very interesting if you are looking for inspiration from very specific examples of how machine learning can be applied. The replies/contributions from the Quora community to this question include such issues as predicting business bankruptcy, cybersecurity, the backdrop and information surrounding a topic you need to make a decision on and which has financial consequences, etc.
This article posted on BBVA API_Market a few months ago describes the positive consequences for banks of the combination of artificial intelligence and big data: enhanced product definition and better fight against fraud. It contains loads of examples, applications and techniques.
And, lastly, a critical article to put your feet back on the ground amidst so much (justified) hype surrounding machine learning and artificial intelligence. Linda Zaikovska-Daukste, CFO/Co-Founder of UX Design Agency, looks critically at the (sometimes) unjustified emergence of interfaces and services based on machine learning derivatives such as chatbots. From the perspective of the User Experience, Linda explains that sometimes these are not the best option to achieve an appropriate client-centric strategy, which she adamantly defends. Interesting opposite view.
A large part of the industry, with years of experience training their teams, designing their strategies and operating their business niches, either voluntarily or under obligation, are having to adapt to new market conditions. One of the most frequent shifts in this industry, including retail and investment banking, is how artificial intelligence can be used as a competitive edge to earn money old- and new-style.
Methods like machine learning and deep learning are helping entities in many different operational fields. Logically, APIs specializing in machine learning and deep learning are the starting point for any transformation. They allow banks to create finalist products that create value for the entity and its customers: they allow extracting important information from Big Data, searching for patterns to tailor offers, price corrections and detecting bank fraud processes.
These days there are application development interfaces that feature natural language processing or image and voice recognition (deep learning) and predictive modeling to make estimates (machine learning). This can be applied in practice: product and customer definition (knowing which services are of interest to each user through customer segmentation); risk management (lending always associated with the possible default); and anti-fraud techniques.
The three key questions in using machine learning for product and service definition and the necessary customer segmentation is where are banking users coming from, where are they now and where are they going. A predictive model must be built which can be interpreted by the operations teams, with the customer at the core of the business logic, and which leads to specific actions. The idea is to define services that are adapted to customer needs and interests, by studying consumer habits and the channels where banking users show the most commitment.
The 2007 global crisis had far-reaching consequences on how financial and investment entities and retailers calculated the risk involved in their business transactions. A recent report by MacKinsey&Company establishes an interesting change in concept: while these days only 15% of bank risk control falls with analytics, by 2025 that percentage will rise to 40%. These changes are always progressive and, as the analysis shows, banks do not need to wait, they can already apply machine learning processes.
This shift of resources in risk management is shown clearly in the following chart, which explains how banks will change structures to assume the new challenges of the new model, based on Big Data technology as machine learning:
Not only will more resources be allocated to early risk detection and not so much to problem solving. This is a strategic decision with a huge impact. Teams will also receive training or external talent will be sought to combat the new forms of bank risk, mainly cyber attacks. Cybersecurity has become a strategic goal for companies and within the financial sector it is a department of great value.
The use of machine learning to prevent fraud is based around methods that can be divided into two general groups: supervised learning and non-supervised learning. In machine learning methods, the machine learns to detect abnormal behavior using a random data subset, which is classified as fraudulent or not. By successively repeating this information processing, the machine improves its predictive capacity and can prevent possible fraud.
The most commonly-used supervised learning methods in this case are supervised neural networks and fuzzy neural networks to prevent both over-the-phone fraud and credit and debit card fraud.
Non-supervised learning, unlike supervised learning, does not include a sample data set that allows machine learning, instead the method aims to identify patterns or similar characteristics to create subgroups for the total data volume. They are common methods like Bayes networks and Markov Hidden Models to establish probabilities and reduce the uncertainty over whether financial fraud has actually been committed.
This is important because, these days, most banks around the world focus their fight against anti-fraud on creating pattern models from subsets of past transactions. Therefore, banks have a low capacity to prevent fraud committed for the first time and in real time. Also, those historical models are not properly up-to-date due to cost reasons. Another important factor is weighing up customer satisfaction: financial entities always carefully consider canceling supposedly fraudulent transactions due to fear of upsetting the customer who, unlike what the predictive model says, performed a legal transaction.
Some financial entities have specialized in solving such problems. Brighterion is one of the fintechs that currently stands out due to its machine learning services to prevent credit card fraud, for example. The company's products combine up to 10 artificial intelligence technologies, allowing the machine to learn, predict and take decisions in real time. It is a cognitive computing platform. Brighterion includes four anti-fraud products:
● iPrevent: the platform can register and learn the behavioral and consumer habits of the owner of any credit cards issued by a bank. The objective: establish red lines which detect possible abnormal behavior when using those cards.
● iDetect: this can detect the violation of personal or security data related to credit cards and irregular transactions.
● iPredict: risk prevention tool for bank credits.
● iComply: uses non-supervised learning processes to detect international money-laundering. The platform receives data from different sources, always in real time, analyzes the data and monitors the money flow between customers and organizations to prevent the laundering.
2017 has been the year in which different technologies have been consolidated, such as React or Node.js, which will continue to be of critical importance in the future. Mobile developments continue to be a stable pillar, but new technologies are arising as the keys to the future after 2018.
Augmented or mixed reality will go beyond the video game sector and become one of the fields with the highest demand for professionals. According to Goldman Sachs, the global VR and AR sector is expected to reach 80 billion dollars by 2025. The sector will offer consumer solutions, tools for the health, general entertainment and video game industries, but also for the creation of additional interfaces.
The launch of mixed reality for Windows 10 and smartphones with ARCore for Android and ARKit for iOS will open new doors and windows for the development of new applications and for the conversion of traditional applications, adapting them to the new paradigms. In addition, new platforms, such as Magic Leap, or dedicated platforms such as Facebook's Oculus or Valve's SteamVR, will have a lot to say in the technologies (hardware and software) that define 2018.
If investors embrace cryptocurrency, such as bitcoin or ethereum, this will open the door to one of the biggest opportunities for innovation in years. Companies like IBM or Microsoft have embraced the technology and are not the only ones. According to Juniper, 57% of the large companies worldwide are considering the implementation of their own solutions based on blockchain.
Therefore, any type of company, not only tech companies, will have to hire or build their own tools. According to IBM, in 2016 there were already 5,000 professionals worldwide working in the blockchain field, but the demand for new programmers, engineers and experts is so high that it exceeds the supply. In particular, Visa is hiring programmers specializing in ethereum, ripple, R3 or bitcoin and its blockchain developments.
The so-called IoT has been here for a long time, but new radio technologies, such as Bluetooth 5 or 5G, will connect almost any element that needs to be connected or improve its performance when connected. The costs are low and there are many opportunities.
The demand for new areas of specialization will be created, from fields like Edge Computing, to the analysis of the data gathered by these devices, including the creation of complex telecommunication priority establishing algorithms. Amazon's solutions, such as Greengrass, or Microsoft's Azure Stack are the pioneers for 2018.
A field as vast as automatic learning allows us to innovate in many different fields, those that were not accessible to the technologies available a few years ago. For instance, the latest report of CB Insights revealed the potential of these technologies in the health sector. From digitizing our personal details to new biotechnology tools and therapies, including the creation of new medical devices.
All of these elements will require teams of multidisciplinary developers and experts in very specialized fields. However, by no means will health be the only sector benefiting from automatic learning. Any other sector with vast amounts of data that can be gathered can be transformed from top to bottom.
With the arrival of PSD2 and open APIs, the financial market will require more consolidated companies to analyze savings and identify potential benefits for their clients. According to Dmitry Budko, founder of Mediant LLC, IA can make the financial services industry "more stable and efficient", mentioning a few examples in other less prominent sectors, such as agriculture, and others in which pressure is higher, such as robotics and personal assistants.
As regards funding, 11.7 billion dollars were raised in 367 agreements with the 100 leading startups of the sector, proof of a constantly changing sector. But that is not all. The main sources of funding come from large tech companies. According to McKinsey, Baidu and Google already invested between 20 and 30 billion dollars in 2016, with 2017 figures yet to be calculated. According to IDC, these figures will reach 57 billion dollars in the cognitive system sector by 2021, with the creation of 30,000 new jobs for highly-qualified professionals.