Data Security in Machine Learning

The Fourth Industrial Revolution is commencing. The fourth installment of the revolution is marked by technology that blurs the line between human and machine and its most sought-after commodity is data. Therefore, data security and privacy are of the utmost concern in the digital economy.

The developments in Machine Learning (ML) are rapid, and the opportunities that this technology creates are wonderous. The rapidity of recent breakthroughs has no precedent, historically and is impacting almost every industry and encounter. Smart technologies are operating in our homes (Alexa, Amazon Echo etc) and in our cars (think Tesla), they are customising our online shopping experiences (Amazon), and soon in store (with facial recognition technology), they are supporting our military and our infrastructure, our agriculture and our financial services. Technology is shaping almost every aspect of human experience. Technology is interacting with, and learning from, the people with which it is intermingling.

But our love of technology also comes at a price. The more of it we adopt, the more we are vulnerable to cyber attacks. ML utilises data to teach its systems, and the better the learning data is, the more accurate the results are. Clean and reliable data is a valuable commodity, but the sheer amount of it makes it also an easy target, but while it must be remembered that cyberattack agents are a constant and real threat, this doesn’t mean that AI technologies are unsafe. They just require a heightened level of data security.

Fighting fire with fire

The best way to protect data privacy and an organisation’s network is by utilising the power of ML. In other words, matching the technology that the opponent is using.

Cybersecurity strategies that employ ML can monitor big data for anomalies. ML models can be trained on “good data” meaning data that shows how the network should be operating. When data does not fit into this ideal, the system is quick to recognise and flag the issue.

Many cyber-attacks begin with a phishing scheme, where sensitive information is stolen through malicious emails or other communication and it is human error that lets the threat agent enter. ML systems can learn from and adapt to, human behaviour. The continuous user feedback that the system gains, helps it to watch for flags and recognise breaches far faster than non-NL security measures.

Limiting the attack surface is a fundamental security practice.

It involves restricting access, employing layered defences and placing smart monitors at each point of weakness in a system.

An attack surface is a summation of all the different points of entry (attack vectors) or unwanted vulnerabilities that a hacker can get data out of, or get malware into, inside a software environ. By shrinking attack surfaces you are looking for a needle in a few pieces of hay instead of inside of a haystack.

ML, when used to support good cybersecurity practices, helps to shrink the attack surface.

Reducing the amount of time an adversary has inside your system to carry out a cyber attack is another good way to reduce the attack surface.

Limiting infiltration time involves the continuous monitoring of an organisation’s data.

Having an automated cybersecurity detection system based on taught ML models means a faster, more efficient response to a breach. At its best, it can interact with the threat agent and lure them away from valuable assets, create a duplicate environment and trap them in it.

Getting educated about cybersecurity

Technology is constantly trying to keep up with the blackhats and vice versa, but the best cyber defence is still educations. According to Varonis, 71% of cyberattacks begin with spear-phishing emails, where the weakest link in defence is always a human. Through proper education on safe cyber-security practices, we can decrease the amount of successful phishing attacks.

Cyber-education should be mandatory for every employee, and should also include executives, who are often targets for “whaling” scams. Whaling is spear phishing geared towards the executive branch who have access to privileged data and corporate funds. A portion of the lessons on cyber hygiene should be teaching employees, through phishing simulation exercises, how to identify indicators of phishing and whaling schemes and what they should do if they receive one.


To truly stay on top of the game we need to bring out the big guns. Educate our employees to work as a line of defence, as well as employ the best cyber security technologies out there. The benefits that Machine Learning will bring to humanity far outweigh the potential detriments to data security and privacy, but it is essential to understand any threats involved. With companies relying on technology more and more, the potential attack surface is growing, but fighting fire with fire is possible when ML technologies are employed to monitor and react to the most sophisticated attacks out there.

How to Prevent Loss of Corporate Knowledge

Human beings have been passing knowledge down to each other since the beginning of time. It’s how we grow, and it’s how we improve our lives. While you would think that it would actually be impossible to lose knowledge now thanks to technology, you’d be wrong.

In fact, it’s a huge problem in today’s business world. It’s estimated that a person will change jobs 12 times within their career. While in the past people stayed with an employer for their entire life, those times are over, and now your employees will likely be jumping around to secure the best offers for themselves.

Training a replacement should be easy, but this is not always the case. This is especially true if one of your employees manages a lot of your company’s information. If they fail to teach that information to your new hires, then your company is going to experience a major informational setback.

In this article, we’re going to talk about how this problem can be avoided. You’ll learn some methods that you can employ in your company to cut back on information loss and how you can make it sting a little less if a key employee changes jobs.

1. Use mentoring to your advantage

In addition to people who will be changing jobs to secure greater benefits for themselves, you also likely have a large group of individuals in your company who will be exiting the workforce entirely. These retirees have spent their lifetime building expertise that has been very useful to your company, and you should make sure that they pass that down before they leave.

Establishing a mentoring program can help them to do that. It also could make your newly hired employees feel more welcome and better prepared for their jobs to learn from someone who has been there before.

This can be accomplished in many ways, and even if your staff doesn’t have the time or the ability to meet in person frequently mentoring can also be accomplished via online mediums like chats.

2. Build a knowledge library

Mentoring is great, but eventually, those people who are acting as mentors will leave. What do you do when they’re not around to mentor your new hires anymore? Well, you could get them to help you build a knowledge library!

This isn’t nearly as intimidating as it sounds, and even something as simple as a wiki could help your newest employees to do their best work. More experienced employees will likely even be happy to assist with this because it can help to silence the thousand questions they may get a day from well-meaning but lost new hires. With a good knowledge library in place, they can simply point them to the wiki for self-service mentoring.

These libraries also have a search function, making it easy to find what you need when you need it, so there’s no need to wade through piles of paperwork to find an answer. Plus, by asking your employees to log important information it’s preserved for the lifetime of your company.

What exactly should go into your knowledge library? That depends on you and your business, but it’s a good bet to include anything that would be important for onboarding. This includes things like work processes, standards to follow, resources and anything else that could help them to become productive members of your team quickly.

3. Better organize your day to day processes

Everyone has their own way of doing things, and that’s not normally an issue while they’re still around. However, it can be a huge problem when they leave, and someone else needs to take over their job. Trying to decipher the messy workspace of a co-worker is a nightmare that leads to a lag in productivity and unnecessary stress for your other employees.

For this reason, it might be a good idea to establish work standards in your office. When you use standards, everyone knows how the system works. If one employee is missing in action either permanently or temporarily, anyone else in their office can immediately jump in to get things back on track.

It also makes it much easier to train new employees when there’s a standard blueprint to follow. A company needs to work together, and that means processes which don’t include everyone will eventually become a hindrance. Using standards preserves company knowledge by requiring that it be recorded in a certain way that makes it easy to use.

4. Take advantage of machine learning

Many people are still not comfortable with machine learning, but it can actually be very useful. While it does take some time to train an AI, once they learn, you can use them forever to perform tasks for you and get work done faster.

This presents an interesting opportunity for your most seasoned employees to teach your AI to perform certain tasks. These processes can continue to run long after that employee retires or seeks greener pastures, not only preserving their knowledge for future use but also automating frustrating workflows at the same time.

Machine learning is obviously not meant for everything, and it does take resources to get started. So, you’ll only want to use machine learning for processes which you’ll need to repeat often. This could include tasks like analysis or data classification which are extremely time-consuming and sometimes even impossible for humans to do.

By using machine learning, you can preserve your data forever and have your AI perform tasks to the same standard as the humans who have trained it. This is not only very convenient, but it also helps your other employees to be more productive because they can use the AI to automate frustrating parts of their responsibilities as well. This frees them up to do other work that’s more important.

In closing, while your employees will eventually make their way out the door, their knowledge does not have to go with them. Your company has likely spent thousands of dollars on this education, and it’s in your best interest to find ways that will allow you to continue to benefit from it.

Dreaming of International Expansion: Regulations in Engineering

Expanding internationally opens up great new opportunities, but especially in the field of engineering, it can also come with a set of new challenges. Many industries are controlled by massive amounts of operating and safety regulations, and being accustomed to a certain set of rules in your home country, does not mean you are covered everywhere else and you could be in for a nasty surprise during your expansion.

This is because every country is different. They all have their own laws and regulations, and failure to comply could cost your company a fortune. Finding yourself halfway through a project, when the authorities come knocking, is an awful position to be in. If the penalties do not kill the project, it could be too expensive to rectify the issues and comply with the government’s demands.

This might effectively be a death sentence, that causes enormous losses for the company. That’s why it’s important to make sure you understand local regulations as well as possible before you begin, but how can you do that?

What should you look out for when taking your company international?

When it comes to complying with international regulations for your engineering company there are a few key areas that you should look out for. Before making the jump to new territories it is important that you research each location thoroughly to make sure your practices will be up to their standards. You may also need to hire a legal aid within the country who is familiar with the regulations in these areas.

Industry regulations

Many safety critical industries are often heavily regulated, and the level of regulation can go deep into the local level. Often an international body tries to harmonise this as much as possible, but we often see differences in country specific regulation due to political, cultural and environmental differences.

Be also sure to check for little nuances in language.

Some regulations approach the end goal by telling the companies exactly how they should do things, while others focus on what the result needs to be, allowing the companies more freedom to decide how to get there.

Environmental regulations

Many countries have extremely difficult environmental standards. This is particularly true if your project will be completed in an area with sensitive and endangered flora and fauna or near protected water sources.

Be sure to pour over the documentation in this area and if it’s not provided make sure to get with someone who can provide this information for you. Ask about protected areas and regulations involved when working within these locations.

Labour standards

If you’ll be hiring labour within the country, make sure you are adhering to all health, safety and wage standards. What’s required here varies greatly by country, but if you’re shown to be a bad actor here, even if you don’t mean to be, you’ll be found out quickly.

Make sure to find out about regulations regarding hours, breaks, time off, etc and make sure that any temporary help you hire is being treated appropriately.

Public safety

If you’re working on a public project, then the standards for them are often times even higher than what is required for private work. Trying to get the facts on these kinds of jobs can be a headache, but when you apply for licensing you should be able to find out what to do here.

Whoever is in charge of this office will likely be able to point you in the correct direction, or you can hire a legal aid with more experience in the country to consult you on these regulations.

Dealing with an information overload

A study by A.T Kearney reported that nearly 80% of companies in highly regulated markets struggle with knowledge management.

That’s a huge number, and they identified that sorting through the information they’ve collected as their biggest struggle.

International regulations produce an enormous amount of paperwork that needs to be completed and documents that need to be sorted through. In fact, there’s so much paperwork that it could actually take your entire team weeks or months to cover all of it to make sure you’re in compliance.

This is not only soul-crushing, but it’s also a huge time drain. To make matters worse, you may also be getting contradicting information from various offices. Large scale engineering projects often require the approval of multiple government entities and it’s foolish to assume that one of them knows what the other is doing.

Assumptions like this will only end up creating trouble for you as they shift the blame to someone else. However, there is a way that you can sort through this information faster and look for inconsistencies.

Modern technologies such as machine learning can be a real lifesaver when it comes to dealing with paperwork in an international expansion.

Help from machine learning?

Machine learning, if you’re not familiar with it, is a type of computer algorithm. It allows you to automate processes and make predictions based on examples it has seen before.

These algorithms can be used for sorting through massive text documents. They can analyse data much faster than humans, find interesting topics and spot risk factors. That means that you can use an algorithm to quickly weed out problems in your plans that could have disastrous consequences.

It can help you check whether you have missed a vital piece of regulatory information in a new country, or if anything has changed in those regulations since last time.

It can also help you find contradicting bits of information coming from different branches and allow you to hammer out these details before the project starts. You can employ the help of your in-country consultant to get concrete answers from officials before proceeding and wasting project funds.

While it does take some time to train an algorithm to perform these tasks for you, once that initial setup is done, you can use that algorithm forever. It will continue to work for you, monitoring important changes, and as you expand you can continue to feed it new data to dissect, sort and manage. It makes managing your international expansion a lot less frustrating.

Should I Teach A Computer To Do It Instead?

Computers are getting smarter and smarter, and some people are understandably confused about the role that artificial intelligence will play in our everyday lives. When you consider the benefits of machine learning it’s best to think about it in terms of how it can make what you already do better. Not as a replacement for an entire workforce.

The truth is that machine learning simply helps you to better manage your staff’s time by completing tedious, time-consuming tasks that humans need not be concerned with. This frees up the humans in your organization to do the important work, leaving the most irritating tasks to machines.

However, there are always advantages and disadvantages to implementing something new and that applies to machine learning as well. In this article, we’re going to talk about the pros and the cons of the process so you can decide whether this is the right move for your company. First though, let’s talk about what machine learning actually does.

What exactly is machine learning anyway?

Machine learning is a type of computer algorithm. While there is, of course, some work to getting them set up to work for your company they are unique in that they can “learn”. Or, actually that they are just very good at predicting outcomes without additional programming. This permits them to use certain rules to make predictions and perform tasks for you.

This can be pretty useful given the right circumstances, and by using machine learning you could actually automate away many of your workforce’s most tedious tasks. Just try to think of the most time-consuming and mind-numbing work your staff is forced to do, and you’ll quickly have some excellent ideas as to how machine learning can free up your valuable work hours.

Why might you want to use machine learning for your organization?

If you’re still not sold on machine learning then how about some real-world examples to help you better understand its many advantages.

Regulatory paperwork and compliance

In many industries, there’s an overwhelming amount of regulatory paperwork which must be completed and even more documentation which must be read and adhered to. Worse yet, this documentation is constantly changing and evolving to meet new challenges.

This makes it very easy for important new additions to be missed by humans, especially if they’ve already spent so many hours poring over the document that they’re having a tough time focusing.

Unfortunately, this is not an excuse that flies with regulatory agencies and any failure to comply could end in very serious legal complications for your organization. What if there was an easier way for you to be sure everything was being taken care of though?

With a properly configured machine learning process you could use a computer algorithm to process all new documentation. It can check for changes or alert you to things that it thinks you should know, freeing up your staff’s valuable time for more important matters.

Recommendations and Error Location

Even if you’re not in an industry with hefty regulations, you can still take advantage of machine learning in order to improve your current business functions.

A properly trained algorithm can pour through all of your documentation, searching it tirelessly for contradictions or potential problems. This can not only save a ton of time versus doing it with human labour, but it could also save your company a lot of money.

Machine learning allows you to catch potential problems before your project gets to the point of no return in its development. This prevents you from making costly mistakes and wasting money on plans which clearly are not going to pan out.

That’s not all though, because an algorithm can also make recommendations for you when it comes to budgeting and efficiency. Computer algorithms are really good at spotting patterns. If the algorithm notices an area where things could be improved, it will help you to identify it.

By using these patterns you can more easily find areas where perhaps your supply chain has been slipping, causing you big losses. Or, maybe you can pinpoint a more efficient way to handle certain business processes, allowing you to save valuable time and resources.

The benefits of using machine learning for these tasks is virtually endless, and they can often do a much better job than humans at tedious, yet low skill tasks. Plus, once your algorithm has been trained it can continue to do the same job over and over without further input from you.

Are there any reasons to not use machine learning?

While it’s possible to do a lot of awesome things with machine learning, there are a few downfalls to the process as well. The largest, of course, is that it can take a substantial amount of time to actually train your algorithm to complete your tasks. (Unless…)

This obviously takes resources and somebody has to develop a model to use for the processes and then test it to make sure that it works. That means that in order for your machine learning process to pay for itself you’ll need to make sure that you’ll be using it regularly.

If the process in question isn’t going to be used very often then it might not actually be worth your time to train an algorithm to do it. In these cases, you might just need to assign infrequent tasks for manual labour completion.

It’s also important to note that an algorithm is only as smart as the person who programmed it.

So, while machine learning can take a lot of work off your plate it needs to be programmed correctly to do it. This includes defining clear goals for the process so it knows what to do when and where.

If the process you’re trying to automate does not allow for this, then machine learning may not be the best choice. Work that requires decisions which may not be absolute is best left to humans because an algorithm will simply not produce the required results. It can only function based on what it knows.

How to Reduce Long Lead Times in Engineering Projects

When it comes to engineering project management, timing is everything. Especially in complex engineering projects such as large-scale infrastructure improvements, commercial builds or custom engineered systems like an industrial crane, marine vessel or even the wing of an airplane.

Engineers and project leads are constantly searching for ways to improve project planning and management processes.

For customers, long lead times mean lost revenue and other opportunities. For vendors, delayed projects turn into wasted internal resources, lost margin and irrecoverable time lost, sometimes even years if they’re truly unlucky. Needless to say, there is no bigger unifying desire across all fields of modern engineering than the need to complete projects at or above expectations, and most certainly on time.

Issues with Long Lead Times

Modern engineering relies heavily on contract text, especially in the procurement stages. And as one would guess, particularly in large-scale engineering projects, evaluating an array of suppliers is a thorough and lengthy process. When multiple offers are submitted, deciding quickly which supplier to go with in order to keep the project moving forward, becomes somewhat daunting.

From a supplier’s point of view, how quickly you respond to a proposal can signal your interest in the project or business and demonstrate how experienced your company is in providing the solution. However, responding swiftly to an RFP can be challenging when tender documents are hundreds of pages long and include regulations, standards and a large amount of technical requirement text. These documents need to be processed and understood quickly, but carefully, in order to gain a competitive edge.

In order to keep up with the demands of today’s markets – an ever-increasing speed of project delivery – engineers and project leads are constantly searching for ways to improve project planning and management processes.

Thanks to advancements in technology and the introduction of automation and digitalization, project managers have the ability to drive down the timelines of projects and ultimately reduce their own project lead times.

Solutions to Help Process Technical Project Data

The complexities of large new engineering projects involve tracking and balancing many moving parts and information. Companies create so much data, yet do not analyze or implement it, creating a lost opportunity to fill the gaps within their project workflows. The value their text data may contain can reveal how to deliver excellent project results and improve their overall efficiency. However, organizing and analyzing all of the previous project data and technical text is a huge task. Project managers can turn these formidable endeavors into opportunities to harness new technologies, thanks to digital advancements like robotics, artificial intelligence, and the internet of things to name a few.

For example, most timeline forecasting done by project managers is based on studying historical data on performed tasks. They’re able to identify how much time should be allocated to each task for successful completion. However, recording and analyzing such data can become an unwieldy and time-consuming assignment. While many project management tools are proficient in storing and collecting historical data, sometimes they fall short with aiding in analytics. Powering the tool with machine learning can not only help automate data collection and analysis but the entire process of predicting realistic timelines can be digitized, which helps project managers achieve gains from the start.

Another application of artificial intelligence (AI) in engineering projects is resource engagement. To ensure projects remain on track, it’s essential that each group in the supplier chain gets the sections relevant to them – but when the written specifications are hundreds of pages long, trying to extract those sections accurately and quickly does not come easily. AI can categorize the requirements based on group function and channel the right information to the relevant engineers, facilitating a more efficient RFP process.

Predictive analytics is another way that artificial intelligence is transforming project management. AI will comb through the specifics of past projects to find out what worked and what didn’t. With this information, predictions about the project can be made that either validates the future outcomes or identifies potential risks and shortages, useful for new project managers or engineers who may be unfamiliar with previous projects.

Of course, AI is not all about automation – deriving actionable insights and finding connections in disparate data is something that even the most trained project manager or engineer could miss. AI structures the data while finding its patterns and inconsistencies, which allows teams to collect insights from dense masses of technical text data, which can be used to improve their project workflows and processes.

Real-World AI Project Management Tools

With technical text data being the heart of large-scale engineering projects, there are a few technologies that can provide some relief from the complexity of long tender documents and safety-critical systems. is an artificial intelligence project management tool that automates repetitive tasks such as following up on meeting minutes, or updating risk registers. It also monitors task prioritisation impacts to improve awareness and organisational agility, and makes sure any built up knowledge can be reused in later projects.

Selko Analytics
Selko Analytics has developed Artificial Intelligence software that can automate the processing and categorizing of text data documents. The deep learning platform for engineering companies can help categorize technical specifications into any function groups and makes sure the right engineers get the relevant information.

This type of technology is useful in modern engineering and can reduce the amount of time and people needed to read through lengthy technical text documents or large projects where stakeholders need to communicate their system needs in long tender documents for custom engineered products.

learn more
The contract intelligence engine analyzes the meaning of not just keywords but of whole sentences, paragraphs, and long text so that the problems of language ambiguity and vocabulary mismatch within and across documents are overcome.

Reduce Long Lead Times with AI

Modern engineering is more technically complex than ever before, and the face of project management is another function that is seeing positive changes thanks to advancing technologies like artificial intelligence and machine learning. In order to meet the market’s demands of delivering engineering projects faster, accurately and efficiently, project teams and leads would do well to leverage emerging AI project management solutions to help them achieve gains from the very beginning.

Exciting Advances in Natural Language Processing

Artificial Intelligence has come a long way since the days of algorithmic chess games. It’s also continuing to evolve all of the time. Weirdly, not that dissimilar to humans…

Alan Turing hypothesized that, if humans have the ability to use what we’ve learned to inform our decisions, then there should be no reason that machines can’t do the same. He even wrote an infamous paper on it.

At the time, however, technology was nowhere near having the capabilities it does now. Computers could only execute programs, meaning they couldn’t actually take anything in.

In the more-than-half-a-century since Turing’s paper, technology has come on leaps and bounds. We now have Alexa, Siri, and Google Assistant sitting in our homes, adjusting the heating at our will.

Chances are, you’re reading this from a small device you’re holding in your hand. Inside of which, there’s a program that’s continuously learning. It’s pretty cool when you think about it.

Artificial intelligence has, indeed, come a long way, but there are still things we need to be working on. Natural language processing (NLP) is one of those things.

In a world closing in on itself, computers and NLP are another way to feel connected. Although, there’s a long way to go until we’re chatting to Alexa the way we would a friend.

Everyday advances in NLP

The English language has so many words that it’s impossible for anyone to put an estimate on how many. Furthermore, a lot of those words has more than one meaning.

Natural language processing is a branch of computer science dedicated to making it possible for computers to process language the way we do. At least, that’s the goal.

Computers can already understand us to a degree. You might’ve even had a bit of fun citing Fight Club with Alexa, but we still have some way to go before computers actually understand the intent of our sentences.

Understanding common sense and reasoning

Unfortunately, AI’s biggest shortcoming is with context. Every day, computer scientists are working to find ways to help computers interact with humans naturally.

One of these ways is with common sense reasoning. By collecting logical assumptions and teaching them to computers, we can make steps toward computers processing language the way that we do.

As mad as it sounds, for all of the words and contexts we know, there are some we take for granted. Take a Mr. Man book, for example.

Mr. Funny is funny, hence his name. We know that, but it’s not something we think about. It’s sort of like saying, “the green light means GO.”

We know that and kids know that, but the world’s most intricate computer systems don’t.

By teaching computers these little snippets of common sense, there’s a good chance we’ll begin to build a more humanistic language.

Sentiment analysis and sincerity

According to the philosopher, Jean-Jacques Rousseau, authenticity is derived from the natural self, whereas inauthenticity is a result of external influences. He might as well have been talking about NP.

Authenticity and sincerity (and their counterparts; fakeness and dishonesty) are fundamental parts of the way we communicate. Although we tend to use tone of voice or body language.

Detecting sentiment in text is a little more difficult. Despite its pitfalls, we’ve found ways to adapt thanks to years of texting, email, and instant messaging.

AI tools have been parsing text for about as long as they’ve been able to. However, they’re not very good when it comes to reading into the sentiment behind the text.

You know when you’ve annoyed your partner when they text you saying, “I’m fine.” Computers, on the other hand, really don’t.

AI automatically takes “I’m fine” as “I’m okay.” Which begs the question: How do people think AI relationships are easy?

A good example of sentiment analysis in text is the buffalo sentence. You know, Buffalo buffalo Buffalo buffalo buffalo buffalo Buffalo buffalo.

It’s completely sincere and grammatically correct. There are no errors and no falseness whatsoever. At the moment, computers and programs don’t understand either of these things.

However, studies and tests are ongoing even as we speak. Check out this Kaggle thread on using algorithms to weed out insincere questions from Quora.

Ongoing research into NLP

Research is always ongoing in the realm of not only NLP but AI in general. In fact, we’ll likely never discover the full capabilities of AI systems.

But that isn’t going to stop us from trying. Some of the research on the go currently is exciting.

Will chatbots be able to identify context and emotion?

As progress continues to be made in common-sense reasoning, sincerity, and sentiment analysis, we’ll get closer to finding this out.

Chatbots already help a lot of websites to run, but they aren’t exactly independent thinkers. Research on whether or not we’ll be able to get them to process context is ongoing.

● Will they be able to remember context they learn through conversation?
● Can they process emotion and respond accordingly? (For example, if a customer is disappointed.)
Will they eventually replace humans in customer service?

Can AI not only generate a joke but understand why it’s funny?

Ask Alexa to tell you a joke and she’ll tell you one that’s been pre-programmed into her. And they’re usually as bad as the jokes in Christmas crackers.

● Will there be a point wherein AI can generate their own jokes? Probably not.

Although, there are machines that have been created specifically to tell jokes. But the jury is out on whether or not that’s a good thing.

Will AI be able to summarise an entire book?

Research on this has been going on for a while now, and we’ve seen impeccable results for text. The next step is to see whether or not AI will eventually be able to summarise books.

● What advances will we see in fields like the medicine and law?
● How will it affect or help with school and university reports?

Will they be able to do all of this unsupervised?

Here therein lies the biggest question of them all.

Catherine Havasi, CEO of Luminoso, says that, without common sense reasoning, it’ll be hard to develop unsupervised systems. Although it’s not stopping anyone from trying.

Unsupervised learning is sort of like teaching yourself an instrument. Computers are given uncategorized, random bits of data and they’re left to work it out by themselves.

But will they?

Problems we might see

Of course, nothing can make progress without possible pitfalls and programs rearing their ugly heads. That’s just the way of progress.

One particular advancement that has people concerned is AI’s potential ability to write. Be it articles or theses, AI journalism is a thing.

But is it really a good thing?

If AI used results to automatically generate text, how will it be consistently 100% plagiarism-free? After all, Copyscape and other tools run on algorithms themselves.

Natural language processing is advancing at rocket speed; we’re going to be seeing it used all the more. Be it in marketing, education, chatbots, or buffalo.

Artificial intelligence opens up an exciting realm of new possibilities. We just have to keep up.

Cutting Through Nonsense in Large Text Documents

Sifting through mountains of irrelevant text just to locate important nuggets of information is as irritating as it is inefficient.

Yet, whether you’re a master’s student putting together your thesis or an engineer trying to categorize technical specifications, you’re going to have to do just that. Worst of all, these kinds of texts run rampant with ambiguities and unclear language making it unusable. Furthermore, they aren’t usually formatted with navigational purposes in mind.

Most of the time, extracting the important information out of text-dense documents is time sensitive, but how can 500 pages be analyzed for specific data in good time?

Improve Your Manual Process

Techniques such as speed reading can aid you tackles these large documents, but there are some faster, out of the box methods, that you can try out and which can help you tackle all that data.

If you’re a lawyer, engineer, economist, or any other highly-educated professional, you’ve had to make your way through undergraduate and graduate programs. One of the main principles of studying and learning is knowing what not to read when trudging through arduous texts such as scholarly and academic journals.

That principle never changes once you’re out of the realms of academia, and since every industry has deadlines you can’t waste time absorbing superfluous information.

Here are some tips on how to get to the meat of larger documents:

1) If a document contains an abstract, read the “findings” and “conclusions” sections first as they usually contain the pertinent information

2) Read the first and last sentence of every paragraph in order to get the gist

3) See if you can find a better-written source with similar information, it may make it easier to work through the more complex text

3) Understand the style format of text you’re reading so you know what to avoid

By mastering these more traditional methods of gathering necessary information, you will be well on your way to better efficiency but harnessing modern technologies can push you the extra mile.

Technology to Rescue

Modern advances in technologies, such as natural language processing, have given rise to a number of intelligent tools to help in different situations. Whether it be thousands of technical specifications, a large pile of CV’s or an uncontrollable amount of customer feedback, we can find tools that can help process the data and give extra insight.

Finding The Right Information In R&D

Collecting a reading list for a new problem takes industrial researchers an average of 3 weeks, with some of the material still being irrelevant for the project. Using artificial intelligence software, these mundane tasks can be sped up by using solutions such as’s exploration tool. builds an interdisciplinary research map based on a problem statement or research paper of your choice, and delivers a precise reading list in less than 2 days.

Accessing Emotional Insights

Many retailers need to understand what’s being said about them on social media. For those who’ve found themselves through various Twitter or Facebook wormholes, social media can present a vacuous hole of worthless text.

Even still, if you don’t assess endless social media posts about your products or services, you won’t be able to optimize your response to issues or fully grasp how your company is performing.

Companies such as Lumoa allow you to track all your customer experience insights from an online platform; finding what drives the customer experience, comparing it with KPI’s and understanding what you are doing well, and what you should be doing better.

A more do-it-yourself solution is also available from Amazon. Amazon Comprehend uses sentiment analysis to computationally determine whether a piece of writing is positive, negative, neutral, or mixed. It can be used as part of a serverless event driven architecture on Amazon Web Services.

Engineering Document Analysis

On the engineering side, tools such as Selko Analytics extract vital information from text-dense engineering specifications. Engineering departments can easily set up their own machine learning model through the tool and pre-process technical text to spot contractual risks, streamline procurement activities, and categorize tasks by groups or architectural levels.

The industry standards demand an influx of data in need of processing, analysis, and implementation, particularly given the complexities of safety and engineering requirements. Time is often of the essence and the engineering specifications are large written documents which need to be processed to get new projects underway.

Selko Analytics’s utilises its intelligent search features to locate items of text adhering to certain groups. Using its user-trained machine learning technology one can properly categorize text after analysing only 100 samples.

Requirements and regulations can be assessed with more efficiency and detail. All results can be verified by a person before export, and any changes made will be learned to produce better results next time.

Solutions Are Out There

Locating the important information in large documents with impenetrable walls of text is frustrating for any professional with a need to get projects and initiatives off the ground.

Thankfully there are plenty of methods to help you through this difficult process. Manual “scanning” methods employed by students are good tips that help with large documentation, but modern technology tools can also be utilised for improved efficiency and productivity.

Text based data can be hard to keep consistent and to navigate, and walls of pointless text can seem frustrating, but there are plenty of solutions abound to cut through the jargon and get right to the point.

What is Natural Language Processing?

Have you ever wondered how services like Siri, Alexa, Cortana, and Google’s assistant work? Perhaps you’re content with a little mystery in your life, but if you want to learn about a new technology that is revolutionizing many industries, read on.

Natural Language Processing (NLP) is becoming increasingly ubiquitous across many devices with new uses emerging frequently. This article gives a quick look at the fundamentals of NLP, what it’s used for and different techniques that make it possible.

What is NLP?

NLP is the branch of artificial intelligence (AI) that is responsible for developing ways for machines to understand human language. Its development has been driven by the growth in big data, machine learning, computational linguistics, computer science and the desire to have more human to machine interaction.

A basic human-computer interaction with NLP may look like this:

• Human speaks to the computer
• The computer captures audio and converts it to text
• Computer processes “translated” text
• The computer converts it back to audio and “speaks” to human

Asimo robot communicating

With advances in technology, NLP is capable of analyzing large volumes of language-based data in a consistent way.

How is NLP Used?

Beyond requesting songs to be played or marked as your favorites, NLP can do much more. Some of the common ways NLP is currently used include:

• Chatbots to automate customer service and ordering
• Improve search results
• Make text processing faster
• Create advertisements
• Provide suggested responses to texts and emails
• Extract information from websites
• Answer complex questions
• Translation and sentiment analysis

At this stage of development, it’s likely that companies have only scratched the surface of what NLP is capable of. As machines and algorithms get more powerful and complex, it’s like that uses of NLP will expand beyond what we can imagine today.

What are Different Techniques of NLP?

While language and communication, in general, rely on syntax, semantics and pragmatics analysis, NLP needs slightly different techniques to carry out its impressive accomplishments.


This technique separates records into different groups or categories based on labels or codes.


Summarization is the process by which NLP can extract a key sentence or develop a short and accurate summary of a longer piece of text.


This technique organizes documents or records within a classification group. It creates clusters within the broader labels.


NLP also relies heavily on extracting data, keywords, keyphrases and other text.


This technique is utilized by search engines frequently as a way to match similar, duplicate or near-duplicate words or phrases. It’s a way NLP can be leveraged to find similarities between different records.

Sentiment Analysis

Sentiment analysis is the technique by which NLP can understand the nuance of and emotion behind what a human is trying to say.

Semantic Analysis

This technique helps machines learn and understand contextual clues humans give when they speak.

Does Your Business Need NLP?

Most likely. If you’re not currently using AI in any capacity, it may be time to catch up.

Many functions within your business – sales, marketing, finance, operations, etc. – could benefit from the adoption of AI and natural language processing.

Selko at Slush 2018 – We’re Excited, and You Should Be Too

Slush 2015, photo by Jussi Hellsten

Selko is excited to be joining Slush 2018, since our experience at last year’s event was nothing short of amazing. Out of thousands of startups, we were chosen as a top 10 finalist in the Slush 100 Showcase (their pitching competition). To say we got the most out of attending Slush 2017 is an understatement.

This time around, as a seasoned attendee and more mature business overall, we can enjoy this year’s event even better than before. Slush started off as a 300 person event a few years ago and has now grown to 20,000 tech heads converging in Helsinki. This year we are focusing on meeting potential partners, industry reps and even possible recruits at the event. That kind of ambiance, energy and overall atmosphere not only brings great talent into one space but also provides a great environment to network and create great contacts.

Tuomas Ritola from Selko pitching on stage, Slush 2017

For More Than Inspiration

Tech events are known for impressive lineups with largely successful names. The great thing about Slush is that not only do they deliver just that, but you also get to dig into real world practices and hands-on advice from these great founders, leaders and startup pioneers. Slush is all about bringing true changemakers to the stage. Here are a few speakers from this year’s lineup:

• Dr. Werner Vogels, CTO at Amazon
• Bill Ready, COO at PayPal
• Julia Hartz, Co-Founder and CEO of Eventbrite
• Katarina Berg, Chief Human Resources Officer at Spotify

The Journey Behind Development

Slush not only knows how to create a great on-stage experience, but off stage there are ample opportunities for companies to share their ideas, products and even their stories. Startup District, Founders Mingle and even Speed Mentoring and Roundtables are just some of the amazing encounters that can happen from Dec 4-5.

And because of all the great contacts and leads from last year’s event, we’ve been in a great position to develop our shiny new MVP (minimum viable product), and we’re excited to show it off in Slush!

Slush 2015, photo by Jussi Hellsten

In Good Company

Of course, it’d be very remiss of us if we didn’t talk about the extensive list of innovative AI and machine learning startups that will be at Slush this year. Once again, we are in great company with a few notable mentions to include:

Malls of Globe – World market offering smart retail platform as a service.
• BEAD – An AI system that analyzes, optimizes and operates a building’s energy management and operations by measuring real-time occupancy.
Something Corporation – A personalized medicine company working on data-driven continuous care for chronic pain management.
Selko Technologies – An AI based software that automates complicated requirements analysis in engineering, that can save years worth of expensive, repetitive expert work

Meet Us There

Slush is known for helping the next wave of tech entrepreneurs take their business to the next level. And with so many like-minded tech fanatics in one phenomenal atmosphere, it’s hard not to.

We’re looking forward to all the great new opportunities to connect and even possibly show off our shiny new MVP. We said it before but we’ll say it again: we’re excited to be joining Slush 2018! Hope to see you there!

Data Overload Providing New Opportunities for the EPCM Industry

Every industry is feeling the changes dealt by advancing technology. Everyone wants projects completed quickly, accurately, sustainably, and cost-efficiently. Engineering, procurement, and construction management companies are expected to meet these desires for other companies, ensuring projects are successfully finished on time and under budget.

To keep up with the demands of other companies, EPCM companies will need to embrace the technology that is the driving force behind change in their industry. Automation and AI tools can help EPCM companies meet the expectations of their clients, completing projects without errors and without missing the deadline.

Advancements in Technology Affecting the Industry

Many industries are beginning to rely on EPCM companies to help them with new projects. With a focus on efficiency, there is a demand for ever-increasing speed of project delivery. Earlier deadlines and faster turnaround requirements are forcing EPCM companies to move faster with less wiggle room in the budget.

Advances in technology are the driving force behind changes in the EPCM industry. Tasks can now be completed faster, with less errors, and sometimes even automated entirely. Robotics, Artificial Intelligence, the internet of things, and additive manufacturing are all impacting how business is conducted and will likely continue to change the business landscape.

For example, drones in a warehouse can monitor and track inventory much more efficiently than people. 3D printing can reduce the time it takes to create products.

Increasing efficiency and decreasing project timelines are goals that a company in any industry could demand. EPCM companies need to expect these demands and work to improve their own efficiency to stay ahead.

Sustainability is a growing concern felt through all industries as well. Construction produces carbon emissions and uses a large amount of resources. Finding a way to reduce emissions and use sustainable materials is a goal of many companies. These goals mean that the previous way of doing business could be changing. For example, cheap material suppliers may no longer be the best option if they are not using sustainable material. This alone can cause upheaval in vendor selection, with contracts and offer documents needing to be read closely for these new demands.

Text Data at the Core

Engineering, procurement, and construction management rely heavily on contract text, especially in procurement. EPCM companies have to go through the process of evaluating a selection of potential suppliers before deciding which ones will be the best choices for the project. They may receive multiple offer documents from different companies, and a decision needs to be made quickly for the project to continue.

These tender documents are often hundreds of pages long. When multiple companies submit offers, they all need to be read quickly and correctly to make the best decisions. If a section is missed or misunderstood, the supplier who is chosen may not really be the best candidate for the job and can result in a regulation or requirement not being met.

Because of the advances in technology, with automation and digitalization driving down the timeline of projects, EPCM companies need to implement automation and digitalization where they can to reduce their own project lengths.

There is so much data that is being created but not being analyzed or implemented to the extent that it could be. Companies that learn how to use the value their data possesses will continue to thrive. For EPCM companies, data can reveal information all through the value chain and can help improve efficiency.

Learning to harness new technologies can also open up new markets. Sending out an offer to a customer that was previous unattainable due to a large barrier to enter, could be minimized by modern tools.

Working across borders or in multiple jurisdictions can also create issues and bring the burden of extra text data to process.

Difference areas may have different regulations or requirements, which create a large amount of extra work before an offer could be made or a project could commence.

All in all, whether it is tenders in procurement, offers to customers or regulatory text, processing this text quickly and accurately is important. However, organizing and analyzing all of the technical text is a huge task. When dealing with very large documents, finding the important information and sharing sections with relevant engineers can be arduous, and in the worst case, vital information can be missed. Luckily, advancements in technology can also help EPCM companies move through project workflows without errors and without breaking the budget.

New Tools, New Opportunities


4castplus allows an entire team – engineers, project managers, project controls, and procurement – to work together and collaborate within a single system. It includes a suite of tools to keep EPCM companies on top of projects, with a full lifecycle procurement system, project controls, and customizable notifications based on system events.

Selko Analytics

Selko Analytics has developed Artificial Intelligence software that can automate the processing and categorizing of text data documents. Machine learning allows the software to be customized to a specific company’s needs, and the software will learn from any changes or corrections a user makes.

This will reduce the amount of time and people needed to read through technical text documents; they simply need to verify the software’s results at the end. Any corrections made are “learned” by the program and will ensure more accurate results with each use.


Aconex improves workflow processes by using one single platform for everyone involved on a project to collaborate. Automated workflows and a project archive will reduce time spent duplicating work, allowing personnel to dedicate their time and resources elsewhere. Integrated search tools allow users to find information quickly, and cost and schedule information ensure projects remain under budget.

There are many tools that can help engineering, procurement, and construction management companies meet the expectations of their clients. Technology is advancing so quickly, EPCM companies will need to embrace it and use technologies such as AI and automation to their advantage and do business faster, more accurately, sustainably, and cost-efficiently.

Learn More