Most of the big data analytics we perform centers around what we expect it to be

“Why is the product unsuccessful? Did we not plan it after going through big data analytics?” asked a company executive. This is not the first company or the first time such confusion has happened. If we take a close look around, most of the data analytics we perform centers around the concept we expect them to be, leading to massive setbacks. Yes, it is true. Even though we praise big data for being the accelerator of every decision, we can’t deny the fact that it can be misleading at times.

Big data is more than just structured and unstructured data. It is seen as a base ingredient of all decision-making processes. For the past two decades, ever since mobile phones came into existence and technology evolved exponentially, It became a critical part of every business operation. Big data in business is a very common substance that executives and employees used to get insights into their market performance. Technology experts are all praises about data, with many touting it as the best thing that has happened to humankind. But the truth is a little twisted. When data is used correctly, it opens the door to double or triple-fold the revenue in minimum time. Unfortunately, it can also be misleading, draining the company’s efforts to go down the gutter. A report by Blazent, an IT intelligence company unravels many findings of big data disadvantages. It shows that around 42% of executives state that misuse of data can impair revenues and 39% said this can be deteriorating for correct decision-making. Henceforth, this article takes you through how big data is misused and what can be done to patch the gap.

Drawing an example from political endeavor

Political circle, especially, presidential elections were heavily relying on big data outcomes. Of course, curiosity didn’t let us be silent. It almost became a custom to know the result through pre-poll analysis. But if we look back at the records, they were not always right. Most recently, the 2016 election that gave Hillary Clinton a 90% chance of victory ended up making Donald Trump the President of the United States. This could either be because of a crack in the data or the data itself was faulty. The big data analytics clearly depicts the fact that human nature as of yet, cannot be reduced to a series of ones and zeros.

Moving on to big data in business and its disadvantages 

Businesses are increasingly relying on big data today. Starting from making simple decisions on marketing and promotion to big ones like where to invest and how to gain more revenues, literally, everything revolves around data. Unfortunately, business executives are unaware that technological innovation is a double-edged sword. If it is not used for good intent, It can wreak havoc.

Datasets are huge and are spread across many disparate locations and diverse forms. Henceforth, business organizations are unaware of whether the data is clean, accurate, manageable, and usable. Besides, some of the data are also manually entered into the system, prompting human errors. While such mismatched data are processed together, it leads to serious negatives and misleading outcomes. However, companies, unaware of the datasets condition take the result as everything and proceed with it.

Businesses are increasingly relying on algorithms to sort company issues. Brian Bergstein of MIT Technology Review suggests that the growing reliance on big data in business is creating a corporate bubble of overconfidence. But why are algorithms unreliable? Even though algorithms are computer-based, they have their own form of risk since they are ‘created by people and they contain interferences and assumptions coded in.’ These coded-in values shape the outputs like computer-generated predictions, recommendations, and simulations.

Finally, one of the biggest setbacks of big data analytics is people’s perceptions. While company executives have a perception on certain products or product developments, the consumers’ viewpoint might vary. But this goes unnoticed when companies focus on delivering their viewpoint to customers without addressing their concerns. Organizations design questions that they want to ask. It is solely on the executive’s perception of what clients needed to answer. They weren’t reflecting on what clients wanted to express. As a result, business takes the wrong path in the name of following big data insights.

What can be done?

Listen to customers. It is the only option to keep away misconceptions. Even though engaging with customers and having a face-to-face or virtual conversation may not be as exciting as compiling big data answers, they reflect on people’s thoughts. When we ask random questions and let them talk, they talk their hearts out and say things that might build the stairs for the organization’s success. For example, Toyota and Adobe are two such companies that go for people’s view than big data decision-making.

The post SOMETIMES, BIG DATA CAN MISLEAD IN TAKING BUSINESS DECISIONS appeared first on Artificial Intelligence.

Deep learning is the current darling of AI. Used by behemoths such as Microsoft, Google and Amazon, it leverages artificial neural networks that “learn” through exposure to immense amounts of data. By immense we mean internet-scale amounts — or billions of documents at a minimum.

If your project draws upon publicly available data, deep learning can be a valuable tool. The same is true if budget isn’t an issue.

But depending on your project, the data you need might be behind a wall, or there simply might not be billions of data points in your dataset. If this is the case, deep learning probably isn’t the solution you need, but you can still draw on machine learning to get results.

Non-Deep-Learning Solutions: High Value And High Efficacy

Let’s assume you work in the pharmaceutical industry. The data volumes in this domain are enormous but are often protected and difficult to get at en masse. It’s also an area with rigorous regulatory requirements that necessitate detailed content classification and auditable results. These factors make it a bad fit for a deep learning solution. But other machine learning approaches can still provide high-value outcomes.

In the pharmaceutical industry, everything is tracked and categorized, so an all-knowing deep learning model isn’t really needed. A more basic type of model (MaxEnt, for example) is sufficient for the task of matching up content with a known taxonomy or identifying new patterns or trends in drug research data.

Top 25 Machine Learning Startups To Watch In 2021 Based On CrunchbaseUnderstanding The Value Of Artificial Intelligence Solutions In Your BusinessUsing Artificial Intelligence To Transform An Industry? Nominations For The 2021 AI 50 List Are Open

Why is this a better solution than deep learning? Because these models are so specific, unlike the more generalized deep learning models, they can be trained on much smaller amounts of data — think hundreds of thousands or millions of documents instead of billions. These models are easier and cheaper to build and are therefore much easier to maintain and update as new data becomes available. Beyond this, the sheer size and hardware demands of a deep learning solution mean that it’s the wrong hammer to use for many of the common problems you encounter in pharma.

A Case In Point: Compendia

Let’s look at the specific case of drug compendia in the pharmaceutical industry. Drug compendia, historically known as “price books,” are essentially summaries of drug information for a specific condition that are shared to pharmacy retail chains, government databases, distributors and EHR databases. They outline which drugs are most preferred by insurers and approved for off-label uses, pricing, and which combinations of drugs do and don’t interact well together.

Compendia are watched incredibly closely because they significantly affect the revenues of pharmaceutical companies. If one drug moves up the list to become the favored drug of providers, this can mean millions of dollars in profits to the drug producer.

The challenge is that compendia are published frequently, and changes aren’t uncommon. This means a significant amount of human time is currently spent tracking changes to compendia and analyzing what these changes mean for a company’s bottom line. Given the number of recognized medical conditions and number of drugs available to treat them, staying on top of these changes is a massive chore for any pharma company.

However, a relatively simple ML differencing model can track and report on changes to compendia over time, significantly reducing the human cost and effort involved while improving the accuracy of the process. Sure, you could solve the problem with a deep learning model, but it wouldn’t be any more accurate and would be dramatically more expensive.

Another example of an ML solution in pharma that doesn’t require internet-scale deep learning comes from work that our company, Lexalytics, has done with Biogen Japan and its Medical Information Department (MID). In this instance, we configured Biogen’s core NLP to identify relevant conditions, ailments, drugs, issues, therapies, and other entities and products within its FAQ and other resources. We used Biogen’s data to train and deploy custom machine learning models into the underlying NLP; the resulting system now understands complex relationships between conditions, ailments, drugs, issues, therapies, and other entities and products. MID operators can type in keywords or exact questions and get back best-fit answers and related resources in just seconds. The system provides more accurate and faster customer service and reduces costs by minimizing the number of calls that need to be escalated to costlier higher-ups within the organization.

Sometimes It Pays To Think Small

While deep learning is the technology du jour, it’s not always the right solution. Deep learning techniques require phenomenal investment as well as access to enormous amounts of data, which, for most business problems, simply isn’t feasible. But for targeted problems with smaller content volumes, less cutting-edge but long-established machine learning techniques or the use of multiple small models can and will improve business outcomes and, with them, bottom lines.

The post Deep Learning? Sometimes It Pays To Go Shallow appeared first on Artificial Intelligence.