In part one of this two-part series, we covered the complexity of the digital identity problem and some early-market solutions. Read on for part two below!

Machine Learning In Digital Identity

Two broad categories of machine learning models are clustering (unsupervised learning) and classification (supervised learning). Each of these has its pros and cons and, predominantly, the advertising ecosystem has embraced clustering to solve the problem of identity.

Clustering

Simply put, clustering is the grouping of similar data into groups, where data in the same group is more closely related to each other than to data in other groups. With this approach, analogous consumer activities based on the association between data fields are grouped. Data fields like IP, user agent, location, time and content consumed are often used to do this.

Even though the approach is scientific and in some cases highly complicated, it is extremely difficult to overcome the flaws of the individual data fields themselves. For instance, IP inherently is an unreliable data field over the long term because the address can change. Identifiers clustered in this model, particularly cookies, are largely unreliable.

Though clustering solves part of the cross-device identity problem, it falls short for a large portion of the device and consumer population.

Classification

Classification, on the other hand, is the process of identifying which label a new observation belongs to, knowing the classification of observations from a fact-based training data set. This method can be used to generate a predictive model to associate multiple advertising identifiers to one consumer or a household.

Using models based on classification for identity resolution involves managing a statistically relevant training set, which has observations from a group of devices known to be linked to the same consumer and/or household.

Having a thorough training set is easier said than done. Imagine collecting training data from your favorite travel website — a consumer might log into the travel website from one or two separate devices to research vacation ideas and deals. It’s highly likely that the same consumer will visit the site anonymously to get quick info like airport information or in-flight movie choices. If that consumer only travels twice each year, the data isn’t being collected on a frequent enough basis to have it included in a thorough training set. Classification models tend to work better for identity resolution if the thoroughness and freshness of the training data can be maintained at all times.

Hybrid Solution For Maximum Accuracy

Not surprisingly, the most effective digital identity solution comes from utilizing a hybrid of these two models. Semi-supervised learning leverages advantages from both clustering and classification algorithms to achieve higher accuracy. At Qualia, we do this by utilizing both algorithms and leveraging a combination of billions of signals from devices and users’ intent-driven activities to create statistically relevant models that can enhance or validate each other.

In semi-supervised learning, a small amount of labeled data is used along with unlabeled data as the training data set for the model. With this approach, the quality, freshness and statistical relevance of the labeled data can be managed, as it is not as cost and resource prohibitive to maintain this labeled data set.

Another advantage in utilizing this model is to perform household identification using classification techniques with labeled data across household devices and then clustering these household devices into individual groups with significantly smaller labeled data sets. An approach like this handles the changes to unlabeled data in a much better way than strictly using classification or clustering techniques.

In short, digital identity management is a complex problem, and there is no one-model-fits-all approach. However, it is one of the most interesting challenges in the advertising ecosystem, encouraging data scientists to innovate and ideate around creative solutions to this ever-changing device landscape.

The post How Machine Learning Truly Applies To Digital Identity appeared first on Artificial Intelligence.

Machine learning is gaining popularity to deal with increasingly complex data and analysis problems. Many projections also point that the highest growth is in India IT spending in software and IT services for 2017. This includes building new digital platforms with Machine Learning at the center. IBM continues to be at the forefront of it all.

‘Machine Learning’ termed by an IBM decades ago has evolved significantly. Today, it enables enterprises to drive critical insights. Businesses are increasingly using machine learning to support advanced analytics across a growing range of industries and initiatives. With India’s focus on digitization, it’s an apt time for organizations to make this transition.

Prasanna Keny, Senior Technical Manager, IBM India says that going forward, machine learning is going to be an imperative and not a choice.“According to a Harvard study 72 percent of organisations are vulnerable to disruption due to digitization and data intelligence. For the respondents, 29 percent of organisations said they are extremely susceptible to market disruption while about 43 percent responded they are significantly susceptible due to the digital intelligence and machine learning used by their competitors. That is how important machine learning, analytics, deep learning or artificial intelligence has become in terms of competition and the ability for an organisation to maintain the leadership position,” said, Keny.

Citing various examples Keny explained the use cases of machine learning. He said it can be used for customer analytics and fraud combating.“A use case of analytics and machine learning is for countering fraud. When we think of fraud, the image that crops up in our mind is online payment fraud. But this can be used in a variety of cases. For instance, we have customers using it for expense fraud, procurement fraud or in case of insurance fraud detection for claims,” added Keny.

However, the whole process or the journey of machine learning involves challenges as well. Some of the challenges are data management where there is unavailability of prior data, the evolving environment and toll sets users, difficulties in collaboration where multiple people are working together and operationalizing.

The post Machine learning is not a choice; it is an imperative: IBM appeared first on Artificial Intelligence.

Looking at the future of marketing and new trends in business, Zachary Harding, local marketing specialist and CEO of Hyperion Equity, has invested in a 10 per cent stake in Blue Dot Data Intelligence Ltd.

Blue Dot is a research and data intelligence company based in Jamaica, which was started a little over two years ago by Larren Peart.

Blue Dot uses biometric technology that tracks eye movement and monitors emotional responses to various stimuli or questions being asked. This ensures that the disconnect between what people say and how they actually feel does not affect the results of the research being done. This way, marketers can understand and target their buyers’ subconscious mind.

“In 2016 elections, locally and internationally, you see where even the most respected research firms got the predictions wrong not because their methods were faulty, but simply because what people say does not necessarily indicate how they feel. Using biometrics, we can know what people are really thinking,” said Peart.

For Harding, the decision to invest in Blue Dot was simple. Having done the appropriate due diligence, Harding deemed that the company’s financials were strong and the growth trajectory positive.

“As a marketer, I always dream about leveraging new technologies to help companies make smarter decisions, both as it relates to their consumers as well as how they do business. The team at Blue Dot was able to show me how easy it is for traditional marketing research firms to get it wrong. It’s not often that I’m surprised, but they were one step ahead of me. The team has the right mindset, and that convinced me that this is something I want to invest in,” said Harding.

New to the marketing world, especially in Jamaica, is the concept of data mining.

“What a lot of people don’t get is the impact that data mining can have on the local economy. As a commodity, data is the new oil, however, many companies either don’t understand the value of their own data or don’t have the expertise to analyse and extract the valuable insights from the data. In some cases, the data they already have can negate the need for them to do market research in the first place. Blue Dot’s emphasis on this area of data mining shows that as a company, they are ready for the future of marketing,” said Harding.

Blue Dot Data Intelligence Ltd has conducted several studies for leading companies in North America, Jamaica, and several other Caribbean islands.

The post Data Intelligence And Biometrics: The Future Of Marketing Research appeared first on Artificial Intelligence.