You are here:  Home  >  Data Education  >  Big Data News, Articles, & Education  >  Big Data Blogs  >  Current Article

Three Tips for Successfully Adopting Machine Learning Technologies

By   /  March 16, 2018  /  No Comments

Click to learn more about author Scott Parker.

It’s no surprise that Machine Learning, Natural Language Processing (NLP) and Cognitive Search technologies are being adopted at high rates. As organizations strive to create value, enhance customer experiences, comply with stringent regulations and differentiate themselves from their competition, they are placing extraordinary demands on their knowledge workers. Frequently, the data and knowledge they need is isolated, segmented and fractured. It’s difficult to surface the right information at the right time and uncover complex patterns in the data.

A well-designed combination of NLP, Machine Learning and search technology enables these organizations to rise to the challenge and leverage enterprise data in unprecedented ways. This technology is effectively powering a new breed of information access that’s faster, more accurate and more thoughtful than ever before. When successfully adopted, organizations benefit by becoming truly “information-driven,” optimizing every employee and customer experience. This transformation is quickly becoming the new competitive advantage since it redefines the way professionals, businesses and industries operate, but how can organizations successfully adopt these technologies?

Align with User Goals

In order to drive adoption of these types of technologies, implementation has to be aligned with every user’s individual needs. While this may seem obvious that the right data needs to be pulled to address specific user needs, this data also has to be presented in an intuitive and timely way that makes it contextual to the user’s goals. The data-driven age is giving way to an information-driven economy where context is critical to surfacing useful insights from data. Meeting user needs means gathering the data, enriching it in the right way, further contextualizing it using the vernacular of not only the industry but the user’s organization and presenting the resultant information back in a way that aligns with the user’s goals.

There is no one size fits all, as each user’s goals and needs will vary. For example, in the customer service realm, customer service representatives (CSRs) increasingly need to become knowledge driven to satisfy and even delight customers. Meanwhile, in the manufacturing or drug development industry, researchers need to become expertise-driven in the sense that they are easily able to connect with experts. It all starts with becoming information-driven.

Start Simple

It is important for organizations to start simple by incorporating context across their enterprise data. This makes it easier for knowledge workers to find and discover the information that is relevant to the task at hand.  Incorporating context means surfacing connections among related data across dispersed repositories as well as acknowledging all of the different ways that language can be expressed, including consideration for acronyms and synonyms.

Within the data, especially the unstructured data, lies the opportunity to add more context through Natural Language Processing (NLP) and artificial reasoning. These techniques, realized by modern technology, can enrich the data and make connections that are meaningful. It becomes less about curating the unstructured data, and more about leveraging it as it exists but in a more valuable way. There are various options for organizations to pursue becoming information-driven, and some of these options should raise a red flag. For example, there is the do-it-yourself approach with open source technology, which organizations should consider only if they are equipped to be a software development shop.

Immerse the Technology into Your Business Environment 

As opposed to immersing the user in technology, the technology should be immersed into the user’s business environment. Technologies such as cognitive search must leverage the vast majority of enterprise data sources, including internal and external data of all types, where it exists, whether on premises or in the cloud. Hence the system must be highly scalable. As opposed to software packages like Salesforce, in which data has to be loaded or input into a single system, an immersive solution leverages the data across dispersed repositories in a secure and scalable way. This in turn streamlines business processes, allowing knowledge workers to spend less time on mundane tasks and more time focusing on important issues.

As organizations fine tune the information-driven approach to their own specific needs, being immersed and starting simple will be key for success. By building on existing knowledge to learn and get smarter over time, this transformation gives organizations an incredible edge in solving tomorrow’s challenges.

About the author

Scott Parker is the Senior Product Marketing Manager at Sinequa. He began his career as a software engineer and systems analyst with Bloomberg BNA. While at BNA, Scott earned a graduate degree in software engineering from Carnegie Mellon University and then went on to become a senior director at Vivisimo where he spearheaded the implementation of the company’s go-to-market strategy. Follow Scott and Sinequa at: Twitter

You might also like...

Thinking Inside the Box: How to Audit an AI

Read More →