Six Trends Shaping Data Science
12 Oct 2021 10:30 am by Mark Dunn
Yep, that’s quintillion with a Q. Data is created by every click, swipe, share, search, and stream. Consider, every minute of 2020, somewhere in the world:
- Facebook users uploaded 147,000 photos and shared 150,000 messages
- Reddit saw more than 479,000 people engage with content
- Instagram users posted over 347,000 stories or clicked on business profile ads 139,000 times
- YouTube users uploaded 500 hours of video
And that’s just social media content, to say nothing of the 28 tracks Spotify added to its music library, the 404,000 hours of video Netflix users streamed and the $1M consumers spent online.
At a town hall co-hosted by the Wharton School of Business and Working Nation in 2019, Tsvi Gal, CTO of infrastructure at Morgan Stanley, remarked: “Data analytics is the oxygen of Wall Street.” And it’s true; global spending on big data and business analytics topped more than $180B in 2019. Seventy-nine percent of enterprise executives agree that companies that do not embrace big data will lose their competitive position, while 84 percent of enterprises have launched advanced analytics and big data initiatives to seize a competitive edge and accelerate their decision-making.
Not only that, but the future of big data is incredibly promising. Data science has been the top job on Glassdoor for the last four years and is estimated to add 11.5 million new jobs by 2026. Demand for these roles is no longer confined to high-tech and software, with massive growth in sectors like education, marketing and manufacturing.
At LexisNexis, we’re seeing six trends rise to the top of the challenges our customers are facing and solutions they’re seeking.
Trend One: Answers
Data scientists want answers. They want analysis. And they want to get there more quickly. They’re seeking solutions that streamline the analytical experience and reduce the time and complexity of curating a data set to answer their research needs.
All-in-one solutions that give users the power to mine data, create visualisations, share and replicate results, and deliver insights in-the-moment are therefore in high demand among data scientists. That’s one of the reasons Nexis has invested so heavily in our new Nexis® DaaS solution, to allow users to move beyond the discovery phase more quickly, in order to get to the more rewarding work of interpreting and sharing their validated findings.
Trend Two: Access
Customers want access to a broader data set without the investment of infrastructure. They want to know the architecture is in place for them to do their work.
The solve for this demand is to ensure the tools data scientists rely on to do their important work offer a familiar, intuitive operating environment that inspires exploration. For instance, Data Lab leverages the best-in-class, open-source web app, Jupyter. The solution comes fully loaded with pre-packaged, open-sourced Python and R language notebook libraries that make it easy for inexperienced researchers to interpret their results, or for users with more research experience to embed their own code. In addition, each user is granted an individual workspace to save and return to searches for further study or refinement.
Trend Three: Assistance
Universities are challenged to prepare their students for the future. With this responsibility in mind, there’s been a move by many universities to form collaborative or innovation hub environments. By placing a greater emphasis on partnerships and projects with industry, universities are giving their students more hands-on training and cooperative work-study opportunities than ever before.
But that also means being able to turn around the aforementioned analysis and access at a faster and more reliable pace—especially when helping novice researchers find, manage and organize their information. That’s where the “New Librarian” comes in. According to Jeffrey Stanton, “The ‘New Librarian,’ is a librarian equipped with data science knowledge,” who can offer helpful resources and drive investigation and learning for data scientists. We view the “New Librarian” as a transformational link between data science and library science.
Trend Four: Assurance
If there’s one word we often hear on repeat from our university customers, it’s this: compliance, compliance, compliance! Universities must be assured that big swaths of data aren't being downloaded or shared in violation of any copyright laws.
This is one of the reasons the Nexis Data Lab solution is so appealing; we manage all of the data within our own environment. Since the full-text data remains with Nexis and isn’t downloaded, our customers don’t have to worry about copyright infringement or data security.
Trend Five: Accommodation
Customers want the flexibility to analyze and verify their findings. We've seen real growth in our data service business, in which customers take our data fields through an API so they can perform their own analysis, look for patterns, compare our data to their data sets and interpret their findings in different ways. Data scientists can expect to see more and more analytical features introduced in the coming years.
This is especially important when it comes to reproduction. Often, students need to share their results with faculty, while faculty need to share theirs with peer reviewers at academic journals. Research reproduction is no problem with Nexis Data Lab: users simply export their analysis, notebook code and document manifest to their laptop and they—or someone else—can resume, replicate and reproduce the exact findings—again, without the need for new investment in additional infrastructure.
Trend Six: Authority
Data science and big data analysis require access to unbiased, unfiltered results—not the kind you get from Google. This is where metadata comes in; it allows users to filter out inaccurate results and biased “noise” before they even get to the results, so they don’t spend valuable time searching for a needle in a haystack.
Through LexisNexis® SmartIndexing Technology™, all Nexis content is enriched with helpful metadata. Users can build searches using index terms—such as subjects, industries, companies, organizations, people and places—to pinpoint relevant results efficiently and accurately, without false hits. This includes article sentiment, so often a critical element in news-based research, which other solutions and the open web do not offer.