Machine learning examples in day to day life.

Posted by

Machine learning is everywhere. It is possible that you are using it one way or another and you don’t know even know about it, read further to find out in how many ways you interact with machines without even thinking about it. In the upcoming posts, I am going to present some day-to-day life machine learning examples which have made our lives simple and secure.

Email Spam filtering:

Motivation: Communication via email has grown as one of the most significant forms of connection. 86% of business professionals prefer to use email when communicating for business purposes. However, unsolicited commercial or bulk emails which are also referred to as spam have become one of the major threats to the email users. Spam e-mails are messages randomly sent to multiple addresses by all kinds of groups, but mostly lazy advertisers and criminals who wish to lead you to phishing sites. Spam accounts for almost 45% of all the emails sent which is about 14.5 billion spam emails every single day. 36% of all spam is some form of advertising. Spam emails can not only be disturbing but also dangerous to customers. Therefore, an effective spam filtering technology is a significant contribution to the sustainability of the cyberspace and to our society.

Traditional Approach:

Merely years ago the majority of the spam could be reliably dealt with by blocking e-mails originating from certain addresses or filtering out messages with certain subject lines. However, then the spammers started to apply several tricky methods to overcome the filtering methods like using a random sender address and/or add random characters to the beginning or the end of the message subject line.

Modern Approach:

Therefore, knowledge engineering and machine learning are the two global approaches used in e-mail filtering. In knowledge engineering approach, a set of rules has to be defined according to which emails are categorized as spam or safe. But even after implementing this method, no significant result was obtained because the rules had to be constantly updated and maintained, which was a waste of time and inconvenient for most users. The machine learning approach is more efficient than the knowledge engineering approach as it does not require any specific set of rules, instead, it requires a set of training samples which consists of pre-classified e-mail messages. A specific algorithm or a set of algorithms are then applied to learn the classification rules from these e-mail messages. Machine learning approach has been widely studied and there are lots of algorithms can be used in e-mail filtering, they include Naïve Bayes, support vector machines, Neural Networks, K-nearest neighbor, Rough
sets and the artificial immune system.

 

email_filter_1.png
EMAIL FILTERING

Online Fraud Detection:

Motivation: Online payment fraud is a reality of the internet age we live in and the numbers are only set to increase with the increasing digital adoption. The number of transactions has increased due to an excess of payment channels – credit/debit cards, smartphones, kiosks. The financial services industry and the industries that involve financial transactions are suffering from fraud-related losses and damages. The most common types of online fraud occur via phishing or spoofing, data theft, and chargeback or friendly fraud.

Traditional Approach:
Over 90% of online fraud detection platforms use transaction rules to direct suspicious transactions through to human review. The rule-based systems often use legacy software that can hardly process the real-time data streams that are critical for the digital space. Surprisingly this traditional approach of using rules or logic statement to query transactions is still used by some banks and payment gateways. The major disadvantage of the traditional process is the occurrence of false positives. This means completely normal customers just looking to make a purchase will go away from your business. The judgment is dependent on individual training and transaction guidelines, which vary depending on the business.

Modern Approach:
There are also subtle and hidden events in user behavior that may not be evident but still, signal possible fraud. Machine learning allows for creating algorithms that process large datasets with many variables and help find these hidden correlations between user behavior and the likelihood of fraudulent actions. It can predict fraud in a large volume of transactions by applying cognitive computing technologies to raw data. Another strength of machine learning systems compared to rule-based ones is faster data processing and less manual work.

Fraud detection process
Fraud Detection Process

 

Online Customer Support:

Motivation:Customer Service is likely one of the most complex and frustrating parts of any business. In any business, it is crucial to provide a high level of customer service. “78% of consumers have bailed on a transaction or not made an intended purchase because of poor service experience.”

Traditional Approach:
Online customer support is handled manually by a call center or a specific management team. The individual handling the call can be from a non-technical background who doesn’t understand the problem the customer is facing or from a technical background from a different department which eventually leads to long calls for the customers and repeating the same problem to 10 different individuals with no guarantee that your issue will get solved in the given time. This is frustrating, inefficient and inconvenient for the customer and they might lose interest in the company’s product and services. In most cases, the reason behind it is that a customer service agent doesn’t know what to do and they hope someone else will know. Furthermore, having a person read every email a company gets to try to figure out what the customer needs and how to deal with it can be a very time-consuming task.

Modern Approach:

LivePerson claims that around half of all customer service interactions are currently highly suitable for bots. Machine learning predicts user future needs based on the history which results in up-selling and cross-selling opportunities. The system even triggers hyper-personalized notifications to CSR(customer service agent) to share with the customer while the customer is still on the call like new products or service offering because this customer searched for that particular keyword in the past. Machine learning for customer service will not only make self-service interfaces more intuitive and economical, but its intelligence will help anticipate specific customer needs learning from their contexts, previous chat history, and preferences.

 

 

One comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s