Have you ever looked over your own notes and struggled to make sense of them? Typically, the more information we try to process, the harder we have to think about how to organize it. When dealing with large amounts of data, we likewise aim to structure it to the end of aiding our decision-making processes. This, in short, is the idea behind information systems.
What Are Information Systems?
Information systems are also known as decision support systems: They aggregate and process an organization’s data, and then output recommendations to their human operators. Thus, you can think of information systems as operating on a human-in-the-loop principle.
As a discipline, Information Systems (IS) sits right at the interface between computer science and business studies. It’s concerned with the design and implementation of systems that maximize an organization’s information flow to facilitate smart, data-driven decisions.
In fact, humans have always been concerned with the transmission of information through specially designed systems. Even today, we have artifacts from long-lost cultures whose meaning we still cannot readily decipher. Information was passed on through writing, works of art, and namely in oral traditions, by use of mnemonic devices like stories and songs.
In those knowledge systems, memory was restricted to spatial resources, such as the storage capacity of the human brain, the walls of a cave or the pages of a book. Over the past decades, those limitations have become less and less pertinent. Nowadays, we can store nearly unlimited amounts of information on our hard drives. This exponential growth in recorded data only increases the demand for intelligent information systems.
Why Are Information Systems Useful?
To extract information from data, we need to structure it in a way that best serves our purpose. But the optimal method won’t always be easy to determine, as different needs may contradict each other. An online warehouse, for example, might want to have a large inventory to deliver goods to their customers quickly — but also keep storage costs at a minimum.
IS scholars research how to best store and process an organization’s data so that humans can make sense of it. The outcome of these business decisions is then analyzed in terms of its usefulness.
As an example, think of recommendation systems. An online retailer will want you to purchase as many items from it as possible. To that end, it’ll study previous customer behavior to understand which offer might entice you to buy more or to spend more time on its website. When you’re browsing through an online store’s catalog, it will then recommend similar items.
But what if you never click on those recommendations, or you click on them but never put anything in your basket? This is your way of telling the system that it’s not good enough and that the online merchant might need to change its decision process.
So in addition to storing an organization’s data and driving its decision-making, an information system also keeps track of those same processes, thus incorporating a self-documenting element. While humans are always part of an information system, certain processes may be automated. That’s the case for most confirmation emails that you receive when making a purchase on the internet.
Where Are Information Systems Used?
Most aspects of modern life are powered by information systems. Since data has been branded “the new oil,” data-based products have vastly increased, which explains the need for ever more sophisticated information systems. Let’s briefly look at three examples of information systems in practice.
At the end of the 19th century, the American Herman Hollerith invented the prototypical information system, which he called the “electrical tabulating machine.” Hollerith’s machine facilitated the 1890 census by automatically reading punch cards with citizens’ data and converting them to tables. The tabulating machine shortened the census evaluation process — performed manually up to that time — from eight years to a mere six months.
Healthcare is a sensitive area by its very nature. Public health officials have to make decisions about an individual’s treatment based on confidential data and other variables. At the same time, the healthcare system is a big apparatus that operates with many different stakeholders. Information systems in the healthcare sector help personnel make informed decisions about a patient’s treatment, taking into account factors such as exam results, the patient’s individual and family history and costs.
Walmart’s Retail Link system is a classic example of an information system. Designed as an inventory management system, it stores and processes detailed data about historical purchases. It then incorporates external information — such as weather forecasts or upcoming holidays — to provide recommendations to the supermarket chain’s suppliers on how to manage their inventory.
As the above examples illustrate, even companies whose product is not based directly on data can benefit from a well-designed information system. That’s why information systems experts are now finding themselves gainfully employed across a vast array of industries.
What Are the Components of an Information System?
Up until now, we’ve been talking about “storing data” as if it were a straightforward task. The truth is that many different data storage techniques exist side-by-side. The choice of database (DB) can depend on which architecture best serves your purpose, or which technology you’re most comfortable with.
The two most popular DB paradigms are SQL and NoSQL. For a long time, SQL was the architecture of choice, based on a simple, rigorous system of interrelated matrices. In an SQL DB, entities and relationships are stored in different tables, each with a predefined schema. Querying them requires a straightforward lookup process.
NoSQL databases, on the other hand, constitute a more flexible architecture. Unlike their relational counterpart, they don’t operate with predefined schemas and allow semistructured and unstructured data in addition to structured data. An example of unstructured data is natural language. NoSQL databases have been growing in popularity in recent times.
Two notable NoSQL storage systems are graph and search databases. A graph DB stores information as an interconnected graph with data points and their relationships represented by nodes and edges, respectively. Search databases facilitate fast full-text searches through the use of inverted indices.
In the era of big data, businesses typically operate with complex data warehouses. These are large centralized databases that often accumulate data points from various sources. Data warehouses usually incorporate some preprocessing pipeline to ensure a certain company standard.
Simply storing your data in a database isn’t enough; you also need a way to read and make sense of data. Every database type comes with a set of query languages. These allow you to pull data according to certain criteria. In addition, a query language lets you write and delete data.
Other than simply reading it, you might want to gain some insights from your data. To that end, engineers may use a collection of machine learning algorithms. These are typically trained on historical data. During training, the algorithms detect patterns in data, which they then use to make predictions about future events.
We’ve previously emphasized the importance of the human factor to an information system. Information systems only exist to help people make decisions. Therefore, user interfaces that are easy to understand and operate are essential elements of information systems. This is of particular importance as the users of information systems are often not tech people themselves, but experts in other fields. Information systems engineers should therefore strive to understand and enhance their users’ experience.
Information system engineers don’t have to be professional coders. But proficiency in programs and computers will certainly help you make much better decisions when it comes to designing information systems in the workplace. Enroll in our expert-taught Introduction to Programming Nanodegree to master the foundations of programming and develop a deeper understanding of computational processes.