Algorithms For Dummies
Book image
Explore Book Buy On Amazon
Algorithms today appear everywhere, and you might not even realize just how much effect they have on your life. Most people realize that online stores and other sales venues rely on algorithms to determine which add-on products to suggest based on previous purchases. However, most people are unaware of the uses of algorithms in medicine, many of which help a doctor decide what diagnosis to provide.

Using sort routines

Without ordered data, most of the world would come to a stop. To use data, you must be able to find it. You can find hundreds of sort algorithms online.

However, the three most common sort routines are Mergesort, Quicksort, and Heapsort because of the superior speed they provide The sort routine that works best for your application depends on the following:

  • What you expect the application to do
  • The kind of data you work with
  • The computing resources you have available
The point is that the capability to sort data into whatever form an application needs to accomplish a task makes the world run, and this capability is changing how the world works.

Looking for things with search routines

As with sort routines, search routines appear in nearly every application of any size today. The applications appear everywhere, even in places that you might not think too much about, such as your car. Finding information quickly is an essential part of daily life. As with sort routines, search routines come in all shapes and sizes. In fact, if anything, there are more search routines than sort routines because search requirements are often more strenuous and complex.

Shaking things up with random numbers

All sorts of things would be a lot less fun without randomness. For example, imagine starting Solitaire and seeing precisely the same game every time you start it. No one would play such a game. Consequently, random number generation is an essential part of the gaming experience. In fact, some algorithms actually require some level of randomness to work properly. You also find that testing works better when using random values in some cases.

The numbers that you obtain from an algorithm are actually pseudo-random, which means that you can potentially predict the next number in a series by knowing the algorithm and the seed value used to generate the number. That's why this information is so closely guarded.

Performing data compression

Data compression affects every aspect of computing today. For example, most graphics, video, and audio files rely on data compression. Without data compression, you couldn't possibly obtain the required level of throughput to make tasks such as streamed movies work.

However, data compression finds even more uses than you might expect. Just about every Database Management System (DBMS) relies on data compression to make data fit in a reasonable amount of space on disk. Cloud computing wouldn't work without data compression because it downloading items from the cloud to local machines would take too long. Even web pages often rely on data compression to get information from one place to another.

Keeping data secret

The concept of keeping data secret isn't new. In fact, it's one of the oldest reasons to use an algorithm of some sort. The word cryptography actually comes from two Greek words: kryptós (hidden or secret) and graphein (writing). In fact, the Greeks were probably the first users of cryptography, and ancient texts report that Julius Caesar used encrypted missives to communicate with his generals. The point is, keeping data secret is one of the longest running battles in history. The moment one party finds a way to keep a secret, someone else finds a way to make the secret public by breaking the cryptography. General uses for computer-driven cryptography today include:
  • Confidentiality: Ensuring that no one can see information exchanged between two parties.
  • Data integrity: Reducing the likelihood that someone or something can change the content of data passed between two parties.
  • Authentication: Determining the identity of one or more parties.
  • Nonrepudiation: Reducing the ability of a party to say he or she didn't commit a particular act.

Changing the data domain

The Fourier Transform and Fast Fourier Transform (FFT) make a huge difference in how applications perceive data. These two algorithms transform data from the frequency domain (how fast a signal oscillates) to the time domain (the time differential between signal changes). In fact, it's impossible to get any sort of computer hardware degree without having spent time working with these two algorithms extensively. Timing is everything.

By knowing how often something changes, you can figure out the time interval between changes and therefore know how long you have to perform a task before a change in state requires that you do something else. These algorithms commonly see use in filters of all sorts. Without the filtering effects of these algorithms, reproducing video and audio faithfully through a streamed connection would be impossible.

Analyzing links

The capability to analyze relationships is something that has made modern computing unique. In fact, the capability to first create a representation of these relationships and then analyze them is the subject of Part III of this book. The whole idea of the web, in fact, is to create connections, and connectivity was a consideration at the start of what has become a worldwide phenomenon. Without the capability to analyze and utilize links, applications such as databases and e-mail wouldn't work. You couldn't communicate well with friends on Facebook.

As the web has matured and people have become more in tune with devices that make connectivity both simpler and ubiquitous, applications such as Facebook and sales sites such as Amazon have made greater use of link analysis to do things like sell you more products.

Spotting data patterns

Data doesn't exist in a vacuum. All sorts of factors affect data, including biases that color how humans perceive data.

Pattern analysis is at the forefront of some of the more amazing uses of computers today. For example, the Viola–Jones object detection framework makes real-time facial recognition possible. This algorithm could enable people to create better security in places like airports where nefarious individuals currently ply their trade. Similar algorithms could help your doctor detect cancers of various sorts long before the cancer is actually visible to the human eye. Earlier detection makes a full recovery a higher probability. The same holds true for all sorts of other medical problems (such as finding bone fractures that are currently too small to see but cause pain nonetheless).

You also find pattern recognition used for more mundane purposes. For example, pattern analysis lets people detect potential traffic problems before they occur. It's also possible to use pattern analysis to help farmers grow more food at a lower cost by applying water and fertilizer only when necessary. The use of pattern recognition can also help move drones around fields so that the farmer becomes more time efficient and can work more land at a lower cost. Without algorithms, these sorts of patterns, which have such a high impact on daily life, can't be recognized.

Dealing with automation and automatic responses

The proportional integral derivative algorithm is quite a mouthful. Just try saying it three times fast! However, it's one of the most important secret algorithms you've never heard about, yet rely on every day. This particular algorithm relies on a control loop feedback mechanism to minimize the error between the desired output signal and the real output signal. You see it used all over the place to control automation and automatic responses. For example, when your car goes into a skid because you break too hard, this algorithm helps ensure that the Automatic Breaking System (ABS) actually works as intended. Otherwise, the ABS could overcompensate and make matters worse.

Just about every form of machinery today uses the proportional integral derivative algorithm. In fact, robotics wouldn't be possible without it. Imagine what would happen to a factory if all of the robots constantly overcompensated for every activity in which they engaged. The resulting chaos would quickly convince owners to stop using machines for any purpose whatsoever.

Creating unique identifiers

It seems as if we're all just a number. Actually, not just one number — lots and lots of numbers. Each of our credit cards has a number, as does our driver license, as does our government identifier, as do all sorts of other businesses and organizations. People actually have to keep lists of all of the numbers because they simply have too many to track. Yet, each of these numbers must identify the person uniquely to some party. Behind all of this uniqueness are various kinds of algorithms.

About This Article

This article is from the book:

About the book authors:

John Mueller has produced 114 books and more than 600 articles on topics ranging from functional programming techniques to working with Amazon Web Services (AWS). Luca Massaron, a Google Developer Expert (GDE),??interprets big data and transforms it into smart data through simple and effective data mining and machine learning techniques.

John Mueller has produced 114 books and more than 600 articles on topics ranging from functional programming techniques to working with Amazon Web Services (AWS). Luca Massaron, a Google Developer Expert (GDE),??interprets big data and transforms it into smart data through simple and effective data mining and machine learning techniques.

This article can be found in the category: