Recent advancements in artificial intelligence, most notably in machine learning and natural language processing (NLP), have introduced automation to a whole range of tasks previously reliant on manual processes, completely transforming how effective and efficient they can be.
As technology and capabilities in the information retrieval and NLP space have accelerated, more and more organizations are rethinking how they source, verify, analyze, and operationalize data, particularly when it comes to unstructured data and Open Source Intelligence (OSINT).
Typically analysts working with news or other content sources rely on highly inefficient manual tasks focused on searching for, reading, verifying, and tagging news articles of interest. Take, for example, an analyst tracking ESG risk, specialising in metals and mining. They require a broad reach across global sources every day, needing to spend time identifying and reading news, as well as recording events that have the potential to matter, via screenshots or some form of clipping. They then need to operationalize the information with their wider group. It’s a highly time consuming, manual workflow.
Automating this process is therefore a key part of intelligence organizations’ digital transformation initiatives — benefiting them by increasing operational effectiveness, lowering operational costs, and reducing reliance on human workers, who (despite best best efforts) have the limitations of being human.
Highly trained human analysts are far from obsolete, however. A hybrid human/machine approach has been proven by AYLIEN customers to provide the best of both worlds: machine collection and processing of news articles at huge scale in real time, which are filtered and validated by human analysts to ensure quality and relevance.
This blog takes a closer look at:
- The challenge of monitoring daily global news at scale
- How an AI-powered News Intelligence solution overcomes the challenge
- The hybrid human/machine approach for intelligence organizations that delivers great results
The challenge of monitoring daily global news at scale
It’s not an overstatement to say that manually collecting, processing, and utilizing the world’s news is impossible. The scale is simply overwhelming, and analysts already spend too much time trying to source news, let alone find, verify, and identify what is relevant.
At best, organizations have been narrowing their scope by focusing on a much smaller sample of tried and trusted sources. This is far from ideal, as intelligence organizations, in both the risk and media industries, are expected by their clients to catch every signal or mention that has the potential to matter.
Developments that augment the manual discovery and investigation process have only been successful up to a point. News aggregation tools, for example, can provide organizations with a huge amount of global news updates everyday. But it can be extremely difficult and time consuming to find the news that is relevant amongst the colossal amount of noise delivered by these solutions. Without a more intelligent solution, analysts may as well be searching for a needle in a haystack.
Embracing the capabilities of AI-powered News Intelligence to overcome the challenge of scale
AYLIEN News API is an AI-powered News Intelligence solution helping intelligence organizations across the world overcome the challenges associated with scale. It enables analysts to automate some previously manual identification and annotation processes, such as tagging and categorization, to make news discovery and investigation faster and easier than ever before. Here’s how it works:
Comprehensive news aggregation
AYLIEN’s extensive content partnerships provide access to real-time and historical news coverage from 80,000 sources in 14 languages from across the world. Every day approximately 1.4 million articles are added to our news archive, providing analysts with a robust and ready-made supply of news articles, with access to a growing historical archive of over 400 million.
AI-powered search capabilities
Access to such a huge amount of news coverage would be overwhelming without a way to easily find what you are looking for. Instead of analysts having to painstakingly annotate articles manually, AYLIEN passes every article through our proprietary NLP engine in real time, enriching each one with 26 data points, producing clean, structured news data. As a result any event and topic can be easily searched for and monitored using AI-powered search filters, which include entities, categories, and event detection.
News delivered to where it’s needed
AYLIEN News API data can be easily integrated into an organization’s analyst-facing app/ product, delivering timely news updates exactly where analysts need them, freeing-up analysts for higher value detection and investigation tasks.
The best of both worlds: a hybrid human/machine approach
Although AI has taken huge strides in recent years, there’s still a way to go before intelligence organizations can rely purely on machines to surface and analyze the information that they need. However, a hybrid human/machine approach has proven to be hugely successful in transforming how analysts work by allowing them to widen and deepen their monitoring capabilities while also blocking out the content and events that may not be relevant.. AYLIEN News API does the heavy lifting of aggregation and tagging news content at scale, which is delivered to analysts’ apps where they can filter for events and topics they are interested in, providing high quality and validated discovery and investigation processes.
For one AYLIEN News API customer, a hybrid machine/human event detection process brought the following benefits and improvements.:
- Operational efficiency: Improvements in analyst efficiency by as much as 40%
- Operational effectiveness: Uplifts in candidate events discovered by as much as 10X
- Coverage expansion: The number of sources monitored daily seeing an increase by 4X