Home General 10 Amazing Examples Of Natural Language Processing

10 Amazing Examples Of Natural Language Processing

Natural Language Processing (NLP)

Natural language processing (NLP) is an increasingly becoming important technology.

This application is helping to power a number of useful, and increasingly common technologies.

From automatic translation or sentence completion to identify insurance fraud and powering chatbots, NLP is increasingly common.

Natural language processing often works alongside natural language generation.

Both are an increasingly common application.

Gartner has predicted that soon, “natural-language generation will be a standard feature of 90 per cent of modern AI and Analytics platforms.”

If you are new to natural language processing this article will explain exactly why it is such a useful application.

We will also see how it is already impacting and improving a number of industries from financial services, healthcare, self-driving cars and many more.

What is Natural Language Processing (NLP)?
What is Natural Language Processing

What is Natural Language Processing?

Natural language processing (NLP) is a form of artificial intelligence that help computer programs understand, interpret, analyze and manipulate human language as it is spoken.

Humans use either spoken or written language to communicate with each other.

Computers communicate via machine code. This is also known as machine language.

At the most, basic level computer communication utilises binary codes.

Over 70 years ago programmers used punch cards to communicate with their machines.

Natural language processing uses technology and big data and sophisticated algorithms to simplify this process.

This application allows humans to easily communicate with computers.

It can be seen in a number of common, every day tools such as Alexa or Siri.

What is Natural Language Processing Used For?

Natural language processing is used in a range of tasks. For example, NLP automatically prevents you from sending an email without the referenced attachment. It can also be used to summarise the meaning of large or complicated documents, a process known as automatic summarization.

Increasingly data is unstructured. This means that it can be difficult, and time-consuming to process and translate into useful information.

Natural language processing and sentiment analysis enable text classification to be carried out.

This sees the algorithm of reading a document and placing it in the correct category.

A common example of this is the spam filter on your email.

Natural language processing is also helpful in analysing large data streams, quickly and efficiently.

A human can take many hours to conduct a similar search, and their returns can be prone to fatigue and error.

This application makes natural language processing useful for analysing everything from medical records to social media posts

Helping Technology Overcome the Language Barrier

Natural language processing and machine translation help to surmount language barriers.

As the amount of online information continues to grow, the ability to easily access information in a foreign language grows in importance.

NLP powered machine translation helps us to access accurate and reliable translations of foreign texts.

Natural language processing is also helping to optimise the process of sentiment analysis.

Natural language processing powered algorithms are capable of understanding the meaning behind a text.

This can help to determine a product or services popularity.

It is also used by TV and production companies to monitor the public reception to new shows.

Natural language processing is also driving Question-Answering systems, as seen in Siri and Google.

This application also helps chatbots and virtual assistants communicate and improve.

These examples show that natural language processing has a number of real-world applications.

Utilising natural language processing effectively enables humans to easily communicate with computer technology.

What are the Steps of Natural Language Processing
What are the Steps of Natural Language Processing

What are the Steps of Natural Language Processing?

The four steps of NLP involve Sentence segmentation, Word tokenization, Part of speech or morphosyntactic tagging and Syntactic or dependency parsing. Some familiar applications of NLP are Machine translations, Chatbots, Optical Character Recognition and Speech Recognition

For natural language processing to function effectively a number of steps must be followed.

These steps are key to natural language processing correctly functioning.

If they are not followed natural language processing systems will struggle to understand the document and may fail.

What are some of the challenges of Natural Language Processing

Properly applied natural language processing is an incredibly effective application.

However, like many technologies, proper implementation faces a number of challenges.

At one-time sentence boundary disambiguation was difficult to achieve.

This requires an application to be intelligent enough to separate paragraphs or walls of text into appropriate sentence units.

While this is now an easier process, it is still critical to natural language processing functioning correctly.

Parts of Speech Understanding

Formal understanding is key for a machine to learn.

In natural language processing applications this means that the system must understand how each word fits into a sentence, paragraph or document.

Parts of Speech tagging tools are key for natural language processing to successfully understand the meaning of a text.

Parts of Speech tags and dependency graphs are also key to helping develop a vocabulary.

While most NLP applications can understand basic sentences, they struggle to deal with sophisticated vocabulary sets.

NLP must also be able to link different parts of vocabulary.

This is key for understanding the wider meaning of a document.

Vector-space based models such as Word2vec, help this process however they can struggle to understand linguistic or semantic vocabulary relationships.

This can lead to difficulties in understanding the context of a text.

Many words with the same spelling can have different meanings.

For example, a bank is entirely different to a riverbank.

The key to bridging some of these difficulties is in building a robust knowledge graph focused on domain specificity.

This allows the system to differentiate between “I enjoy working in a bank” and “I enjoy being near a riverbank”

Enhancing methods with probabilistic approaches is key in helping the NLP algorithm to derive context.

Similar difficulties can be encountered with semantic understanding and in identifying pronouns or named entities.

These challenges must be surmounted for NLP to become a perfectly robust system.

MORE – Computer Vision Applications in 10 Industries

Natural Language Processing Applications in Finance

Natural language processing can help banks to evaluate customers creditworthiness.

Thanks to NLP, these assessments can be accurately carried out with minimal financial records.

This application sees natural language processing algorithms analysing other information such as social media activity or the applicant’s geolocation.

Lenddo uses natural language processing to accurately assign credit scores to applicants. It uses traditional and 'non' traditonal data as such social media networks to determine a score.
Lenddo uses NLP to accurately assign credit scores to applicants. It uses traditional and ‘non’ traditional data as such social media networks to determine a score.

Analysing these factors allows NLP to gauge the applicant’s habits and their relationships.

These related factors help to create an accurate credit score.

Lenddo is a Singapore based technology solutions company.

They have developed the Lenddo application.

This uses NLP to accurately assign credit scores to applicants.

In partnership with FICO, an analytics software firm, Lenddo applications are already operating in India.

Here, according to the World Bank, around half of Indians do not receive the right level of financial support.

Lenddo applications are helping lenders better assess applicants, meaning that millions of more people are able to safely and responsibly access credit.

Lenddo applications are also currently in use in Mexico, the Philippines and Indonesia.

MORE: Equifax Launches Machine Learning Powered Credit Scoring System

Personalising the Banking Experience

Natural language processing is also helping banks to personalise their services.

London based Personetics have used natural language processing to develop the Assist chatbot.

This is designed to help personalised banking. Assist can be integrated into websites, messaging platforms and apps.

It uses the customer’s previous interactions to comprehend queries and respond to requests such as changing passwords.

Royal Bank of Canada's Chatbot utilises Natural language processing to personalise customer experience
Royal Bank of Canada’s Chatbot utilises NLP to personalise the customer’s experience

RBC's Chatbot utilises using a form of AI called NLP to reccomend ways for customers to save money
RBC’s Chatbot utilises using a form of AI called NLP to recommend ways for customers to save money

The Royal Bank of Canada has already integrated Assist into its mobile banking app.

Personetics reported a 20% increase in the app’s usage after integration with Assist.

Similarly, SAS has developed the SAS Platform.

This uses natural language processing to analyse customer feedback and improve customer service.

A cloud solution, the SAS Platform uses tools such as text miner and contextual analysis.

This allows algorithms to understand and sort data found in customer feedback forms.

These insights are presented in the form of dashboard notifications, helping the bank to create a personal connection with a customer.

The Royal Bank of Scotland is just one financial institution to use the SAS Platform.

SAS's Contextual Analysis uses Natural language processing and machine learning to analyse customer feedback
SAS’s Contextual Analysis uses NLP and machine learning to analyse customer feedback

MORE: 10 Applications of Machine Learning in Finance

Optimizing Contract Processes

With the help of Python programming language, natural language processing is helping organisations to quickly process contracts.

JPMorgan Chase has developed COIN (short for Contract Intelligence), a text mining application that can read and parse commercial loan contracts.

COIN is able to process documents, highlighting and extracting certain words or phrases.

When done manually this is a repetitive, time-consuming task that is often prone to human error.

This application of NLP is reportedly saving the company 360,000 hours every year.

This application of NLP helps to speed up time-consuming tasks.

Consequently, skilled employees are able to concentrate their time and efforts on more complex or valuable tasks.

JPMorgan Chase is aware that automation and sophisticated tools have endless possibilities in the banking sector.

Jamie Dimon CEO (r) of JPMorgan Chase has been a big supporter of AI and its possiblities to transform banking
Jamie Dimon CEO (r) of JPMorgan Chase has been a big supporter of AI and its possibilities to transform banking

Automation also means that the search process can help JPMorgan Chase identify relevant customer information that human searchers may have missed.

As with other applications of NLP, this allows the company to gain a better understanding of their customers.

MORE: JPMorgan Chase Hires Another High-Profile AI Expert

MORE: JPMorgan Creates an AI-Powered Virtual Assistant to Support its Clients

Optimising Healthcare Provision with NLP

Natural Language Processing is helping the healthcare sector to optimise administration.

This is helping to counteract the growing problem of physician burnout.

A New England Journal of Medicine study revealed that 83% of respondents were concerned about physician burnout.

Over half the respondents also believed that automating administrative tasks would decrease the workload on physicians.

Much of this administration concerns the constant reviewing and updating of Electronic Health Records.

NLP tools will allow physicians to dictate automatically to the EHR during patient consultations.

They will also be able to access the records quicker. NLP automation would not only improve efficiency it also allows practitioners to spend more time interacting with their patients.

WellSpan Health in Pennsylvania is using NLP voice-based dictation tools in this way.

R. Hal Baker, MD is Chief Information Officer and Senior VP of Clinical Improvement at WellSpan.

In his opinion, this is a “much more cooperative approach – not to mention a more efficient one. I can talk to both the record and the patient at the same time, so I don’t have to walk out of the room and recount the entire visit again at some later time. That lets me spend a greater percentage of my time in the patient’s presence.”

MORE: Top 10 Ways Artificial Intelligence is Impacting Healthcare

Improving Patient Literacy

Natural language processing is also helping to improve patient understanding.

Increasingly patients are using portals to access their health records. This is done with the aim of helping the patient make informed lifestyle choices.

However, the benefit is only realised if the patient is able to understand their records.

A poll conducted in 2016 revealed 15% of patients struggled to understand their EHR record.

Limited patient understanding can also prolong consultations.

Practitioners may find they have to stop and explain terms and results in detail.

Natural language processing can be applied to EHR data.

In 2017 researchers used natural language processing tools to match medical terms to clinical documents and lay-language counterparts.

This was done with the aim of easing patient understanding.

A similar study saw researchers developing natural language processing tools to link medical terms to simple definitions.

When the patient logs into the portal to view their EHR they are able to easily decode complicated terms and results.

This leads to the patient developing a better understanding of their condition.

While both studies delivered interesting results, a system has yet to be developed that can be used in real-world scenarios.

Identifying Patients Most in Need of Care

External factors, such as housing instability, as well as mental health disorders can make it difficult for patients to stick to treatment regimes.

Patients failing to follow recommended treatment plans can often require further treatment.

These patients can also often incur more care costs during the course of their lifetime.

Massachusetts General Hospital have been using NLP and machine learning to structure data before passing on recommended treatment plans to patients
Massachusetts General Hospital have been using NLP and machine learning to structure data before passing on recommended treatment plans to patients

Natural language processing, as well as machine learning tools, can make it easier for the social determinants of a patient’s health to be recorded.

As this information often comes in the form of unstructured data it can be difficult to access.

Again NLP and machine learning, can improve access, ordering and structuring the data before presenting it in a useful way.

Researchers based at the Massachusetts General Hospital have been using NLP and machine learning in this way.

Similarly, natural language processing can help to improve the care of patients with behavioural issues.

Beacon Health Options is a behavioural health management service provider.

They are using NLP and machine learning to mine unstructured data with the aim of identifying patients most at risk of falling through the cracks in the healthcare system.

Dr Emma Stanton is Beacon Health Options’ Associate Chief Medical Officer.

She revealed that “Our goal is to move from being a reactive model that solely looks at what has happened historically to being a much more predictive, proactive, and targeted service provider,”

Stanton sees this application as a way of helping “an incredibly vulnerable segment” of society.

MORE: Artificial Intelligence in Medicine – Top 10 Applications

Tesla is one of the many car manufacturers to develop self driving technology utilising a range of AI including natural language processing
Tesla is one of the many car manufacturers to develop self-driving technology utilising a range of AI including natural language processing

Natural learning processing in Developing Self-driving Vehicles

Vehicle manufacturers are working toward developing autonomous vehicles.

For autonomy to be achieved, AI and sophisticated tools such as natural language processing must be harnessed.

In addition to enabling autonomy, NLP will also allow commuters to engage with media, such as video games, whilst travelling in a vehicle.

One of the keys to any new technology becoming a success is its ability to develop trust with the consumer.

Natural language processing will be key in the process of drivers learning to trust autonomous vehicles.

BMW's Intelligent Personal Assistant allows you to give instructions is an example of natural language processing in self driving cars.
BMW’s Intelligent Personal Assistant allows you to give instructions is an example of natural language processing in self-driving cars.

A key to a fully automatic vehicle will be the ability to verbally communicate with the car.

In other words, the passenger will simply get in the car and instead of driving or programming a Saatnav will simply tell the car where to go.

NLP and AI algorithms will be key to achieving this level of communication and understanding.

In the process of achieving this, BMW has developed an in-car personal assistant.

Similar to other smart assistants, this is a voice-operated application.

As well as helping the driver locate a nearby gas station, or identify a short cut this assistant also controls in-car features such as temperature control or music choice.

As this application develops, alongside other smart driving solutions NLP will be key to features such as the virtual valet.

MORE: World’s Top 33 Companies Working on Self Driving Cars

Tesla uses natural language processing in its Enhanced Summon feature which will allow to you pick up your car with your voice.
Tesla uses natural language processing in its Enhanced Summon feature which will allow to you pick up your car with your voice.

In this application instead of walking to your car, it will come and pick you up.

Tesla is developing this application as the Summon feature.

Similarly, natural language processing will enable the vehicle to provide an interactive experience.

This application is particularly useful in an emergency scenario, such as the driver becoming incapacitated.

Here 5G and NLP will allow a third party to access the vehicle and steer it to a safe stop.

In-car personal assistants are yet to achieve this level of capability.

However, the speed at which technology is developing means that it is not far away.

Messenger Chatbots are Helping Companies Connect with Customers

Facebook Messenger bot is increasingly being used by businesses as a way of connecting with customers.

Natural language processing tools are key to this development of functionality.

Uber, in 2015, joined the many companies to launch a Facebook Messenger bot.

NLP powered Uber messenger bot
NLP powered Uber messenger bot

The Uber bot is designed to make ordering a car a quick and easy process.

This process is optimised further if Messenger has access to the destination address.

In 2016 Mastercard also launched a Facebook Messenger compatible chatbot.

Acting as a virtual assistant, the Mastercard bot helps users track their spending habits.

It also alerts users to potential benefits on their card.

This bot allows users to easily manage their finances without the need to adapt to a new app.

This optimisation of connectivity is proving popular.

A survey recently revealed that 73% of people prefer Messengers live chat to using the phone or email to communicate.

Statistics like this are too important to be ignored.

Especially when businesses also learn that every month Facebook Messenger has 1.2 billion active users.

Marriott, the international hotel chain, uses a Facebook Messenger chatbot to let customers alter reservations or redeem points.

Meanwhile, stationers, Staples use their bot to send customers personalised updates and shipping notifications.

MORE: 10 of the Most Powerful Chatbots Today

Marketing Chatbots 

Makeup brand Sephora also has a Facebook Messenger chatbot.

The Sephora chatbot allows customers to easily book appointments.

Integration with the Sephora virtual artist chatbot also helps customers to identify products, such as specific lipstick shades.

This helps to streamline and speed up the purchasing process.

Similarly, Pizza Hut’s Messenger chatbot allows you to quickly order a pizza.

It also reminds you of past and favourite purchases and highlights current deals you may be interested in.

Pizza Hut’s Messenger chatbot
Pizza Hut’s Messenger chatbot

MORE: Dominos Using AI to Make the Perfect Pizza

Personalising communication between consumers and brands is clearly the way forward.

By developing a presence in Facebook Messenger brands can communicate in a casual manner with customers.

Helping to build the brand, and brand loyalty without it ever feeling like a hard sell.

Natural language processing allows for the automation of customer communication.

This helps a brand to build a presence and maintain commercial awareness.
Automation also enables company employees to focus on more high-value tasks.

MORE: Facial Recognition: All you Need to Know

Amazon’s Echo is a good example of consumer product using natural language processing
Amazon’s Echo is a good example of consumer product using natural language processing

Enabling Personal Assistants

In recent years digital personal assistants, such as Alexa have become increasingly common.

The success of these bots relies heavily on leveraging natural language processing and generation tools.

Not only is NLP key to digital assistants functioning, but it is also helping these tools to evolve and improve over time.

Every time that Alexa or Siri responds incorrectly it uses the data derived from its response to improve and respond correctly the next time the question is asked.

Alexa is capable of carrying out over 70,000 skills.

NLP and machine learning has been key to this evolution happening so quickly.

It has also benefited from Amazon’s decision to give developers free access to AVS.

This allows outside developers to build tools and features for Alexa.

There are currently over 28,000 smart home devices that can integrate with Alexa.

This number is only going to increase.

More than just a tool of convenience, Alexa like Siri is a real-life application of artificial intelligence.

Sentiment Analysis and Monitoring Social Media Effectively with NLP

Knowing what people are saying about you or your products is key to maintaining a good reputation.

If your product has a bad reputation it could lead to the demise of your business.

A BrightLocal survey revealed that 92% of customers read online reviews before making a purchase.

86% of these customers will decide not to make the purchase is they find a significant amount of negative reviews.

Natural language processing tools such as the Wonderboard by Wonderflow gather and analyse customer feedback.

This allows businesses to see how products or services are received.

By monitoring, customer response businesses are able to respond to problems and maintain a good reputation.

Sentiment Analysis and Social Media

Natural language processing allows businesses to easily monitor social media.

This is done through the process of sentiment analysis.

Sentiment analysis helps to determine the attitude and intent of the writer.

Natural language processing also helps with coreference resolution.

This is key to a computer interpreting the text correctly.

For example, social media site Twitter is often deluged with posts discussing TV programs.

Using NLP driver text analytics to monitor viewer reaction on social media helps a production company to see how storylines and characters are being received.

Sprout Social uses NLP tools to monitor social media activity surrounding a brand
Sprout Social uses NLP tools to monitor social media activity surrounding a brand’s reputation.

Text analytics can also be used to attract advertisers.

Instead of just pitching a viewing figure, the TV company can show advertisers the exact demographic that a program has.

This enables advertisers to better see how a TV program can match to a brand.

Social media listening tools, such as Sprout Social, are looking to harness this potential source of customer feedback.

Sprout Social uses NLP tools to monitor social media activity surrounding a brand.

Increasingly major organisations, such as General Motors, are using social media to improve their reputation and product.

Predicting and Managing Risk with Natural learning processing

Natural language processing can also help companies to predict and manage risk.

One company delivering solutions powered by NLP is London based Kortical.

Their Kore platform is designed to help financial institutions develop AI systems to forecast risk.

An unnamed investment bank has reportedly used Kortical to optimise and speed up their trading risk prediction process.

Kore platform is designed to help financial institutions develop Natural language processing systems to forecast risk.
Kore platform is designed to help financial institutions develop AI systems to forecast risk.

Kore was able to easily analyse and categorise the nearly 5 million overnight trades that the institution needed to process.

Kotical claims their prediction patterns improved the time this analysis took by 30%.

Amongst the other organisations using these systems, the NHS is using Kortical to reduce costs and waste.

Reducing Risk in the Health Industry

For the financial sector NLPs ability to reduce risk and improve risk models may prove invaluable.

As well as identifying potential fraud or insider trading, NLP can also help to ensure regulatory compliance.

This application is also useful for the health industry, particularly in areas such as pharmacovigilance.

This refers to studies conducted after a drug has been marketed.

By continuing to monitor the use of a drug, the company is able to gather information on its side effects.

This largely involves looking for adverse drug events in patients electronic health records.

By using NLP tools companies are able to easily monitor health records as well as social media platforms to identify slight trends and patterns.

This is commonly done by searching for named entity recognition and relation detection.

Manual searches can be time-consuming, repetitive and prone to human error.

NLP allows for named entity recognition, as well as relation detection to take place in real-time with near-perfect accuracy.

Natural language processing allows companies to better manage and monitor operational risks.

This in turn not only makes processes safer and more robust, but it also helps to cut costs and errors.

Sinuhé Arroyo Founder of Taiger uses Natural language processing driven solutions for the insurance industry.
Sinuhé Arroyo Founder of Taiger uses NLP driven solutions for the insurance industry.

NLP to Help Optimise Insurance Claims Handling

Processing insurance claims can be a time-consuming process.

Agents have to check through claims, cross-match information with policies and frequently interact with the claimant.

This is often a long, drawn-out process. However, natural language processing can be used to help speed up this task.

IBMs text mining software Watson Explorer and Taiger are both NLP driven solutions to the insurance industry.

Both solutions are capable of speeding up and optimizing claims processing.

The IBM Watson Explorer is able to comb through masses of both structured and unstructured data with minimal error.

This application is useful for processing insurance claims.

It can also be used by customer service personnel when searching for the right information.

MORE: Aon Partners with AI Insurance Startup Zesty.ai to Transform Underwriting

WestPac used IBM Watson to increase customer interactions from 40% to 92% of customers.
Westpac Bank used IBM Watson to increase customer interactions from 40% to 92% of customers.

Delivering Noticeable Results

IBM claims that Watson Explorer has helped an unnamed leading insurance provider to organise their data into an accessible database.

This helped call centre agents working for the company to easily access and process information relating to insurance claims.

Introducing Watson Explorer helped cut claim processing times from around 2 days to around 10 minutes.

This helped the company’s 14,000 agents save, on average, around 3 seconds per call.

Watson Explorer has been implemented by a number of leading organisations such as Westpac Bank.

Speeding up claims processing, with the use of natural language processing, helps customer claims to be resolved more quickly.

This helps to improve customer satisfaction.

It also frees up agents to focus on more complicated or high-value tasks.

Similarly, Taigers software is designed to allow insurance companies the ability to automate claims processing systems.

This virtual assistant can search a claim, extracting the relevant information and providing insurance agents with the right information.

Speeding up access to the right information also negates the need for agents to constantly question customers.

Again helping to save time and optimise the process.

Taigers software is versatile and scalable.

It can be used to process a range of claims from lost items to car accident claims or house damage.

MORE: Allianz Rolls Out AI in Handling Claims

Fraud Detection Applications

Insurance fraud is a costly expense.

It also serves to drive up the price of insurance premiums, making it harder for honest people to get cover.

Natural language processing is proving useful in helping insurance companies to detect potential instances of fraud.

One company working to implement NLP solutions in this area is Azati.

They have developed an NLP driven machine learning system that is proving impressively accurate when detecting causes of fraud.

They are not the only company making developments in this area.

Meanwhile, Health Fidelity is providing natural language processing software to identify cases of fraud in the healthcare sector.

Health Fidelity uses Natural language processing to identify cases of fraud in the healthcare sector
Health Fidelity uses NLP to identify find structure in unstructured data.

Health Fidelity’s HF Reveal NLP is a natural language processing engine.

It is able to complete a range of functions from modelling risk management to processing unstructured data.

This application can be used to process written notes such as clinical documents or patient referrals.

Another feature of the software is the terminology engine.

This can be used to search and identify important phrases or words and can be adapted to that particular task.

During the training of this machine learning NLP model, it would have learnt to not only identify relevant information on a claims form but also when that information is likely to be fraudulent.

Natural Language Processing to Help Fight Crime

Natural language processing software can help to fight crime and provide cybersecurity analytics.

NLP is able to quickly analyse and derive useful intelligence from both structured and unstructured data sets.

This application is increasingly important as the amount of unstructured data produced continues to grow.

Sintelix understands relationships between words (NLP in action) and recognizes entities, delivering entity and relationship extraction capabilities at high accuracy in multiple languages.
Sintelix understands relationships between words (NLP in action) and recognizes entities, delivering entity and relationship extraction capabilities at high accuracy in multiple languages.

Sintelix uses NLP powered software to help authorities find patterns and relationships in crime data
Sintelix uses NLP powered software to help authorities find patterns and relationships in crime data

Sintelix is one company providing solutions in this area.

Sintelix utilises natural language processing software and algorithms to harvest and extract text or data from both structured and unstructured sources.

This application is able to accurately understand the relationships between words as well as recognising entities and relationships.

It also operated in a number of languages.

By continuing to develop and integrate NLP and other smart solutions on smart devices presents intelligence professionals with more information and opportunity.

Identifying Fake Crimes

Cardiff University and Charles III University of Madrid researchers have developed an AI system named VeriPol.

Utilising intelligent algorithms and NLP, VeriPol is able to identify fake crime and false theft claims.

VeriPol uses NLP to analyse written statements, identifying patterns that are commonly associated with false claims.

This development is essentially a lie detector test for the written word. Computer scientists behind this software claim that is able to operate with 91% accuracy.

Developers believe that VeriPol can help the police to decide where best their resources should be invested.

It may also prove useful in identifying members of the public who are filing false claims.

This is a crime in itself, and costs law enforcement agencies valuable time and money.

VeriPol is already being used by law enforcement agencies in Spain.

Dr Camacho-Collados said, “Ultimately we hope that by showing that automatic detection is possible it will deter people from lying to the police in the first instance.”

Natural Language Processing in Only Going to Increase in Functionality and Importance

Natural language processing is an increasingly common intelligent application.

From helping people understand documents to construct robust risk prediction and fraud detection models, NLP is playing a key role.

As the amount of data, particularly unstructured data, that we produce continues to grow, NLP will be key to classifying, understanding and using it.

From crime detection to virtual assistants and smart cars as technology continues to advance, NLP is set to play a vital role.

MORE – Computer Vision Applications in 10 Industries

Subscribe to our newsletter

Signup today for free and be the first to get notified on the latest news and insights on artificial intelligence

KC Cheung
KC Cheung
KC Cheung has over 18 years experience in the technology industry including media, payments, and software and has a keen interest in artificial intelligence, machine learning, deep learning, neural networks and its applications in business. Over the years he has worked with some of the leading technology companies, building and growing dynamic teams in a fast moving international environment.
- Advertisment -

MOST POPULAR

Computer Vision Applications in 10 Industries

Computer vision, or abbreviated to CV, is an increasingly important technology in the field of artificial intelligence. Those...