Nand Kishor is the Product Manager of House of Bots. After finishing his studies in computer science, he ideated & re-launched Real Estate Business Intelligence Tool, where he created one of the leading Business Intelligence Tool for property price analysis in 2012. He also writes, research and sharing knowledge about Artificial Intelligence (AI), Machine Learning (ML), Data Science, Big Data, Python Language etc... ...Full Bio
Nand Kishor is the Product Manager of House of Bots. After finishing his studies in computer science, he ideated & re-launched Real Estate Business Intelligence Tool, where he created one of the leading Business Intelligence Tool for property price analysis in 2012. He also writes, research and sharing knowledge about Artificial Intelligence (AI), Machine Learning (ML), Data Science, Big Data, Python Language etc...
Data science is the big draw in business schools
517 days ago
7 Effective Methods for Fitting a Liner
527 days ago
3 Thoughts on Why Deep Learning Works So Well
527 days ago
3 million at risk from the rise of robots
527 days ago
Top 10 Hot Artificial Intelligence (AI) Technologies
11 Tech Leaders Share The Real Truth About Artificial Intelligence (And What Really Matters)
A lot of times, artificial intelligence is marketed as the quick fix to solve all of your business problems. Businesses should be wary, however: AI is not a quick fix, and marketers shouldn't buy into the jargon right off the bat. And a quick fix cannot, by definition, be a customized one. Above all, businesses need solutions that are tailored to their specific needs, not something that's one-size-fits all.
Before exploring the realm of AI, businesses should first ask themselves if and why they need AI-powered solutions. If a business doesn't have an issue that AI can solve, such as acting on mountains of data with human-like ability to reason, they would be smart to go beyond the solution-of-the-week mentality and find a customizable solution that's right for their business.
Further, understand that garbage in equals garbage out. When approaching new technologies built on the foundation of your internal or external data, businesses can find themselves in a â??walk before you run' situation. They must take care of basic needs like data hygiene before exploring the technologies to be built on top of that data."
Artificial Intelligence in commerce is often misunderstood and overhyped. Retailers should see AI as another tool to enhance and improve the customer experience. One key area AI enhances the customer experience is with personalization. There is no doubt retailers will start to see an increased demand for highly personalized commerce experiences from consumers. According to Episerver's Reimagining Commerce report, more than a third of consumers feel brands do a poor job of personalizing the customer experience -- which will ultimately mean missed revenue for retailers. A subset of AI, machine-learning, leverages real-time consumer behavior and historical order data to automatically recommend products, promotions, and content that are unique to each individual consumer. What's exciting is we're now seeing tech-innovative retailers use AI-powered personalization experiences beyond web and mobile, and extending into physical stores such as shoppable mirrors or recommended assortments for associates to show customers."
The practical uses of AI in retail / eCommerce is centered around solving 3 basic pain points: 1. helping customers find what they are looking for on and offline, 2. keeping the "right" amount of inventory at the "right" place in the "right" product mix, 3. real time pricing adjustments to stay competitive. The 1st pain point is addressed by AI enabled search, sometimes called "insight engines" and the use of chatbots to manage the interaction. The 2nd pain point is addressed through the use of AI to provide dynamic inventory forecasting or order velocity forecasting to predict needed inventory levels and locations, and dynamic products assortments to optimize the product merchandising plan. The 3rd pain point is managed through a dynamic pricing engine that constantly monitors pricing and dynamically adjusts to keep the product price optimized. In all three cases AI provides the ability to take large and dynamic data sets and product an output that gets better as the AI engine "learns" and adjusts from dynamic feedback."
AI is already being leveraged in very practical ways -- to streamline tasks, infuse relevance and personalization into people-machine communications, and augment human intelligence and action. Some of the brands we're working with are already harnessing a variety of AI elements (machine learning, natural language processing, predictive, speech analysis, etc.) in feedback scenarios. Allowing the technology to dynamically adjust follow up questions based on what customers are saying results in very human-like conversation that also net significantly better data.
AI is also being applied to understand what customers are saying on social media posts, videos, and phone conversations -- in real and near-real time -- and then automatically alert multiple places across the business on everything from supply chain issues to legal-safety emergencies. Another practical and current business application of AI is in mining massive sets of every data type imaginable, instantly and continuously. The AI "watches" for patterns in human conversations that indicate something unusual is happening. Business users can pick up the thread and easily understand what, where and most importantly why the anomaly is occurring and then take intelligent action. This type of AI might be more accurately referred to as SAI -- super artificial intelligence -- because it can peruse a quantity of incredibly complex and divergent data in seconds and minutes, something even an army of the smartest, fastest human analysts couldn't accomplish. By both automating the impossible and mundane, and giving humans a massive boost in their ability to exercise more informed judgement -- the area where robots fall down -- AI can make businesses smarter, faster and even more human."
The rise of chatbots is a crucial AI application for businesses to embrace, but the tech world is further away from a real AI chatbot than the hype would lead you to believe. It's important for brands to focus first on the problem-solving aspect of this first, rather than the AI aspect, ultimately still utilizing a human-in-the-loop with the goal of making them more efficient.One thing we've heard from social customer service teams is that, when they're fielding a high volume of inquiries day in and day out, they don't have insight into which customer issues or questions they should prioritize when building a chatbot. That offers an opportunity to use machine learning to cluster similar messages together, ultimately surfacing most frequently discussed inquiries or topics. A human would be able to see those clusters and intuitively know which issues they should prioritize for their chatbot. This type of sorting, analyzing and qualifying using machine learning in tandem with human insight is where I see the biggest opportunity for brands."
According to the Strike Social report, "The Big Picture: 2017 YouTube Advertising Benchmark Report," the retail industry view rate for YouTube ads is 76 percent below the all-industry average. With 18- to 34-year-olds being the most influenced by YouTube, the retail industry needs to figure out how to better engage this group to drive sales. That's where AI comes in. By using machine-learning algorithms, advertisers can identify new segments within the 18-34-year-old range - both demographically (e.g. gender) and related to the platform's targeting (e.g. different affinity audiences) - and optimize a campaign towards them. This method allows us to help brands find different ways to approach any struggling segments to achieve better performance."
Increasingly, retailers are applying AI tech to the in-store experience by using robots and in-store chatbots to increase engagement and help consumers find exactly what they need within the store. Today, these experiences are largely centered on using robots to accept and understand consumer voice commands to help navigate a consumer to the right area of the store for the products of interest, and retailers are even starting to test robots picking and packing customer orders on behalf of the customer. Additionally, ecommerce has long leveraged AI for product recommendations - according to McKinsey, 35 percent of Amazon purchases come from recommendations based on algorithms that correlate past customer purchases, searched product, and what others have purchased to determine what should be shown to any given consumer. And it's only a matter of time (likely measured in days or weeks) until voice-controlled assistants like Amazon's Alexa go beyond just order execution to start making product recommendations."
The sheer volume of available data and the proliferation of accessible Machine Learning libraries like Google's TensorFlow and Spark MLLib have given developers far more options to build software that can actually learn and react. And as barriers to AI are lowered, it's opening doors for retailers and other organizations that typically don't have significant R&D budgets. For example, product recommendations can now be driven by more intelligent models that consider an individual at a point in their journey. Perhaps learning from recent transactional history can predict a life event and offer associated recommendations, or analysis of recent shipping delays can trigger a real-time interaction or offer to prevent attrition or a support request. Just as Netflix aims to predict what users want to watch as they fire up their app, retailers need to find opportunities to learn from individual behaviors and other relevant data to provide meaningful engagements through the appropriate channels."
Artificial intelligence has the potential to transform product data throughout the entire eCommerce ecosystem, from the depths of the supply chain to customer-facing content. Over time, this technology will give manufacturers unprecedented insight into the products and criteria that are most important to consumers, which they can use to optimize sourcing and strategic product development. On the consumer side, we will see more robust information targeted to each individual's specific interests, instead of the â??one-size-fits all' model. On both sides, product data will be informative, personalized and actionable for every party involved."