Case Study – Acquire Media Corporation
Acquire Media’s customers have a variety of expectations and output formats for their newsfeeds. This is partly because of the variety of classification systems that are commonly used for classifying news articles, such as the Global Industry Classification Standard (GICS); the Industry Classification Benchmark (ICB – created by Dow Jones Indexes and the FTSE Group); the International Press Telecommunications Council (IPTC) NewsCodes; and the North American Industry Classification System (NAICS). Moreover, Acquire Media has developed four proprietary news classification systems to handle customer expectations for classification that weren’t being covered by the more commonly used systems.
An additional classification system entered the picture in the summer of 2007, when Acquire Media bought NewsEdge, a large but underutilized news classification system. According to Stewart Clark, Acquire Media’s vice president of sales for publishing, “NewsEdge users have had to rely on antiquated technology. The next phase will be to move the customers onto a more robust and flexible platform.”
To operate with the extreme efficiency and accuracy required in the news aggregation business, this new platform would require a thorough integration of the various classification codes into a single filtering system.
From Acquire Media’s website:
“We are an advanced technology company devoted to delivering text news and multimedia with guaranteed reliability in formats that make content instantly usable and searchable. We serve the news, digital publishing and financial services communities, which depend on the speed of our delivery service, our reliability, our 24×7 support, plus our ability to handle multiple formats and delivery options.
“Our specialty is news and other time-sensitive content – acquiring it from publishers and distributing it to corporate servers, websites, and end users with near-zero latency and pinpoint filtering. We are generally regarded as the industry’s most technically-sophisticated media delivery service for aggregated content.
“We provide syndication and delivery services to some of the most respected publishers in the world including Dow Jones News Service, PR Newswire, Business Wire, The New York Times Syndicate, McClatchy Interactive, McClatchy Tribune, CanWest, Canada NewsWire, Briefing.com, The Press Association, Market News International, Tribune Media, United Features, and LA Times-Washington Post News Service.
“We provide inbound content management services for major players in the financial services market place: JP Morgan Chase, Schwab CyberTrader, Credit Suisse First Boston, Wall Street Source, Mergent, Thomson, NASDAQ, FactSet, SNL Financial, Fidelity, AlphaTrade, and Wall Street on Demand.
“We deliver more than 500,000 news articles a day, over the public Internet, with near-zero latency.”
Access Innovations taxonomists used MAIstro software to create a master thesaurus for unifying the different classification systems. The basis for the thesaurus was Access Innovations’ NewsIndexer thesaurus, to which the IPTC NewsCodes were already mapped; additionally, the rulebase associated with NewsIndexer already contained complex rules for automated indexing and filtering of the NewsIndexer terms and corresponding IPTC codes.
Each classification system was organized differently. The hierarchical organization of the thesaurus maintained much of the structure of the original NewsIndexer thesaurus, with modifications to better accommodate added terms. The industrial section of the thesaurus was based largely on the ICB and GICS systems. The terms and corresponding codes of each system were mapped to existing or newly added thesaurus terms.
Each concept is represented by a single thesaurus term, but the terminologies of the different systems are still easily viewable, as well as usable in data processing. This enables two-way newsfeeds, with input and output possible in either the thesaurus terms or in any of the various classification systems that had been mapped to the thesaurus.
The new thesaurus provides Acquire Media with possibly the only thesaurus that unifies several different news classification systems. This serves as a basis for in-house thesaurus and rulebuilding development at Acquire Media, to align the thesaurus and the automated indexing rules with Acquire Media’s software and with the specific content they handle. The final result is a world class news aggregation system.