Using machine learning, AI and analytics to get contextual meaning out of any content whether text, audio or video, SPIRE claims to have built world’s first Context Intelligence Platform offered as PaaS which can be applied to any functional domain in any industry- such as finance, SCM, legal, education, defense etc.
The context intelligence platform understands the data, making it far more meaningful and useful with far-reaching efficiency and cost saving. It has proven numbers of 95% accuracy in contextual search and 80% accuracy in demand-supply mapping.
Used in the current scenario in HR industry, the solutions offered by SPIRE is making the recruiters life easy by eliminating human bias and streamlining the hiring process.
To understand more about it, AIM interacted with Neeraj Sanan, CMO, SPIRE technologies, where he opened up about contextual analytics, use of machine learning in bringing contextual meaning out of the text, solutions offered by SPIRE and more. Read the complete interview as below:
[dropcap size=”2″]AIM[/dropcap]Analytics India Magazine: Could you tell us about SPIRE and its work in Contextual search and Intelligence?
[dropcap size=”2″]NS[/dropcap]Neeraj Sanan: Artificial Intelligence as a concept is not something which has a uniformly accepted definition. If you ask me, I would tell it coming out of the policy paper of the White House which was filed in late 2016, which says that AI is a system which has the capability to act like a human, think like a human, act rationally and think rationally. So AI is a system which can mimic the behavior of human being.
Now talking about the contextual analytics on textual data, the opportunity where SPIRE is trying to build a business stems from the fact that as internet gets converted into Internet of Everything (IoE), machines are developing a capability to communicate with each other. And the data which is coming out of it is huge. While the amount of data being churned is increasingly humongous, amongst that let’s say only 10% is being analyzed. This is because most of the data which is generated over internet is unstructured, not numeric and its mostly text. And as we move to rich media, a large part of it still continues to be text.
This textual data has a lot of information which is going unutilized because no one is tapping onto it. That’s the business we are trying to do in terms of textual data capture. And why contextual analytics? If we extrapolate in an enterprise level, each enterprise is different and may have different level of choices and standards. Good for one may not be good for other. These contexts need to be taken care of when you are analyzing textual data and textual data analytics is all about analyzing qualitative words.
In a nutshell, when you are dealing with text data, you are trying to get meaning out of abstract things on qualitative parameters and they have to be analyzed qualitatively and quantitatively.
AIM: How is Spire using machine learning to bring a contextual meaning out of the text?
NS: The context is of transient nature and the meaning of right might keep changing with time. The transient nature of context would mean that any technology or individual who is trying to analyze it needs to be transient as well over a period of time. This is where machine learning comes into picture.
For instance, the concept of calling has changed from landline to mobile phones to Skype but when it changed we do not know. The same way context keeps changing and humans keep adapting to it. Machines have to be told to adapt. They need to continue to learn from data and fine tuning their context. SPIRE adds value to customers by doing textual/data analysis contextual to our customer.
If the context of my customer is transient, I need to have a technology which is transient. What we do is we build data, we build prescriptive analytics for people to use it. And if the decisions are not accurate, we keep self correcting ourselves to become more and more accurate. So the more we use machine learning, the better it is, as it has the capability of self correcting itself.
AIM: How is it being used in businesses? Would you like to highlight a few use cases?
NS: SPIRE has built a platform for analytics and have instantiated it through our first product called SPIRE TalentSHIP, which is essentially a contextual talent science and AI platform for HR. Our customers get an autonomous HR which builds talent ecosystem for them. Our customer realizes the value of our product by converting their whole human resource function into an exceedingly efficient and quantifiable function.
Today HR in lot of companies is abstract and not very quantifiable. Organizations suffer from huge inefficiencies which typically summarizes as- if you look around for a job or a suitable candidate, there is a dearth in both. How can both be scarce? Either the skills are scarce or the jobs are scarce. So there is an inefficiency somewhere. SPIRE is working on correcting this inefficiency by creating a technology that can understand and read textual data and then comprehend it contextually to the organizations.
AIM: What are the other solutions offered by Spire?
NS: Spire TalentSHIP is not just about recruitment but is a suite of five products which are cloud based technologies and each of these five products is converting different parts of the talent supply value chain into an autonomous process. Right from finding the right candidate to marketing them, to fitting them onto the right role and upskilling them, HR has a lot to do. And the products from Spire TalentSHIP has offerings for this entire cycle starting from hiring to a candidate’s exit from the company. Starting from social tools to deployment tools to management tools and an overarching control dashboard for the CXOs, Spire TalentSHIP offers every solution to convert HR into an autonomous department. The aim is to upskill the HR department so that they too have a significant say in the business.
For instance, suppose there is an HR who goes to the boss and tells last year we had 100 recruits out of which 80 are average and 20 are top performers whereas this year we had 80 out of which 60 are high performers. How do I prove which one is better? This fitness to the job and their performance can be quantified by the tools SPIRE is offering so that HRs can prove to their CEOs how much profits they are adding to the company.
AIM: How do you think has the use of AI increased in the last few years, especially in HR?
NS: AI is a buzzword right now which started somewhere early last year. However, its adoption is a long way to go. HR technology market is today globally in an excess of 17bn USD and this is just the HR tech market . If you see, most of these HR technologies are at least 7 years old and are laptop based technologies. Most of them don’t even work on mobile phones. All of these factors have led to the adaption of newer technologies such as AI in this field.
However, AI being the buzziest word may not in reality have seen the biggest business adoption till now. It has a lot of potential but the penetration is very low. Will it happen? I hope it does. And with the kind of transformation it promises, people are ready to adopt it.
Try deep learning using MATLAB