Blog

Machine Learning…with context

Synetec Machine Learning

Topic identification, also known as pattern recognition, is a critical component in providing context for data analysis if you want to achieve real, valuable business insights.

In our previous articles we’ve discussed the reasons why undertaking data analysis can be crucial for companies to improve their growth and operations.  You can read our intro article here, and our second article in this series here which focuses on using analysis as part of your compliance and regulatory processes.  As we’ve mentioned before, valuable business insights can be achieved with NLP (Natural Language Processing) and ML (Machine Learning) analysis tools to identify patterns, behaviours and sentiments within customer interactions.  There are three important components that make up these solutions with the first one being topic identification.

Topic identification is carried out by NLP and ML methods to identify the different subjects that are referenced in human interactions.  It is the most important of the three components and is particularly complex to implement due to the conversational nature of human interaction.  As you’ll know, when communicating a variety of subjects can be, and usually are, discussed in a single phone call or email.  Being able to correctly identify the right topic at the right moment within a conversation is absolutely critical because without this information, any analysis will be lacking context and become irrelevant.  It’s all about context. An additional benefit of topic identification is that inaccuracies within transcription records will have minimal impact on the overall results because while specific words or sentences can sometimes not achieve 100% accuracy, the ‘gist’ and context relevant to the topic is what really provides the insight.

So how do you measure the accuracy of topic identification to ensure you’re getting the right information?  Firstly, you must be testing a pre-processed data set so that you know what information it contains.  Using a manually processed data set will allow you to identify the accuracy levels and determine which specific algorithms will give you the best results for that data set (it can vary based upon factors such as the communication medium).  Secondly, you need to evaluate the two aspects that are used to measure accuracy.  These two aspects are precision and recall.  Precision is the accuracy of identifying the correct topic and recall is how many times the topic has been correctly identified within the conversation.

Below you can see the testing that we carried out here at Synetec in relation to the detection of advice.  We tested 6 different methods which were Decision Tree, Logistic Regression, Support Vector Machines, Random Forest, XGBoost and Deep Learning (MLP).  We measured the precision and recall levels for each method separately with their own benchmarks.  For precision, we selected a benchmark of 60% and for recall we selected a benchmark of 75%.

We chose these benchmarks because it’s more important we’re able to identify a higher level of instances where advice is detected even if this also produces more false results.  If we increase the precision accuracy for correctly identifying each instance this would lower our recall accuracy, resulting in less instances where advice is detected overall.  We identified (as illustrated in the graph below) that for our purposes Logistic Regression is the best performer when the two aspects are combined.

It’s crucial for achieving relevant insights that the method you choose will provide the accuracy levels you’ve benchmarked.  These benchmarks will vary based on the volume and type of data you want to analyse, and the purpose of your analysis.  As you can see below, another benefit of the Logistic Regression method is that the precision and recall measurements are variable with an increase in one effectively lowering the accuracy of the other.  This means that when identifying different topics we can adjust the accuracy levels to align with an organisation’s analysis goals.

As you can see there are many important considerations to make when looking at topic identification.  Skipping large portions of relevant data or providing too many irrelevant results will mean any business insights become far less valuable and possibly inaccurate.  Topic identification provides the context for your data analysis, regardless of its purpose, so it’s critical to invest in getting this right.

Stay tuned to read about behaviour identification and sentiment analysis which we’ll be tackling in our future articles!

 

If you would like further advice on topic identification or how machine learning and data analysis can provide a meaningful contribution please contact us.

 

Synetec is an Agile solutions provider with expertise in diverse development technologies, such as Angular, the .Net Framework, SQL Server and other cloud friendly data stores. We are certified and have successfully delivered projects across different cloud technology stacks such as Microsoft Azure and AWS, delivering integration and development solutions since 2000.

We work with a number of the UK’s most respected financial institutions to deliver a range of innovative solutions. We have expertise in working with both established businesses as well as start-ups and extreme growth businesses.

Advantages of Angular 5

 

Synetec Angular 5

Version 5.0.0 of Angular, pentagonal-donut is here! This major release contains an abundance of new features.  The aim is to continue the Angular team’s focus on making it smaller, faster, and easier to use. But even though we’re still in awe of the release, it’s business as usual for the Angular team.

Angular 5 – Don’t know about you but we’re very excited that it’s finally here! It contains a lot of new features:

 

Build Optimizer

The build optimizer —a tool included in their CLI for making users’ bundles smaller using the team’s semantic understanding of users’ Angular application— has a double job:

  • the Angular team was able to mark parts of users’ application as pure, which improves the tree shaking provided by the existing tools, removing additional parts of users’ application that aren’t needed.
  • the tool removes Angular decorators from users’ application’s runtime code. Decorators are used by the compiler, and aren’t needed at runtime and can be removed.

The aim of these jobs is to decrease the size of your JavaScript bundles, and increase the boot speed of your application for your users.

 

Angular Universal State Transfer API and DOM Support

Angular Universal focuses on helping us developers to perform server-side rendering (SSR) of Angular applications. By rendering your Angular applications on the server and then bootstrapping on top of the generated HTML, you can add support for scrapers and crawlers that don’t support JavaScript, and you can increase the perceived performance of your applications.

There’s also ServerTransferStateModule and the corresponding BrowserTransferStateModule, which allows you to generate information as part of your rendering with platform-server, and then transfer it to the client side so that this information does not need to be regenerated. The team has also welcomed domino to platform-server, which means that more DOM manipulations are supported out of the box within server-side contexts.

 

Compiler Improvements

The Angular compiler has been improved to support incremental compilation — This provides faster rebuilds, especially for production builds and builds with AOT. The Angular team has also added features to the Decorators and made it possible to ship smaller bundles by removing whitespace.

HttpClient

Developers appreciated the new HttpClient (introduced in version 4.3)  so the team is now recommending HttpClient for all applications, and deprecating the previous @angular/http library.

To update to HttpClient, you’ll need to replace HttpModule with HttpClientModule from @angular/common/http in each of your modules, inject the HttpClient service, and remove any map(res => res.json()) calls, which are no longer needed.

CLI v1.5

Starting with v1.5 of the Angular CLI, the team added support for Angular v5.0.0 and will generate v5 projects by default. The build optimizer has been turned on by default, so developers can now benefit from smaller bundles.

They’ve also updated the way .tsconfig files are used to follow TypeScript standards more strictly. By default, the CLI configures TypeScript without files or include sections, so most developers won’t be affected by this.

 

New Router Lifecycle Events

There’s a new lifecycle events to the router which allows developers to track the cycle of the router from the start of running guards through to completion of activation.

.

 

Contact us if you need advice on your development requirements.

 

Synetec is an Agile solutions provider with expertise in diverse development technologies, such as Angular, the .Net Framework, SQL Server and other cloud friendly data stores. We are certified and have successfully delivered projects across different cloud technology stacks such as Microsoft Azure and AWS, delivering integration and development solutions since 2000.

We work with a number of the UK’s most respected financial institutions to deliver a range of innovative solutions. We have expertise in working with both established businesses as well as start-ups and extreme growth businesses.

Agility and Innovation with AWS

Developing a leadership culture at AWS

We recently attended the AWS Summit here in London and particularly enjoyed their talks on tackling innovation and their hiring culture. If you haven’t attended an AWS Summit before we’d highly recommend it! AWS, Amazon’s Web Services platform hold summits around the globe each year that are always full of new and thought-provoking information.

The AWS approach to innovation has three main pillars which are; customer obsessed, long-term thinking and a willingness to fail. This is a great way to look at innovating and we believe that of all the pillars, being customer-obsessed is absolutely crucial for any company to excel in their sector.

The talk also spoke about executing innovation and that to do it well there’s four areas that you have to develop to be successful. These were culture, architecture, organisation and mechanism. While they are all equally important, we thought the two most interesting concepts from this were culture and organisation. AWS really focuses on hiring ‘builders’, looking at the personalities of the people they hire, not just their technical abilities.

The second was organisation and the specific way they put teams together to improve agility. They call them ‘2-pizza teams’, meaning that if you were to order two pizzas it would be enough food to feed that team. The reasoning behind it is that small, single-threaded teams can produce more speed and agility. By decentralising authority, every team owns what they’re creating which empowers them to think big and produce results.

We finished off the day by attending a talk on the habits of Alexa consumers, who, where and how is Alexa being used within households. We found it quite surprising that only 33.5% of the households surveyed use their Alexa in the living room and the most used application is setting a timer.

Emerging Alexa Habits at the AWS Summit in London 2018

Overall, the AWS Summit was absolutely fantastic to attend as always and we’ve come away with some great information to share with the team.

 

Synetec is an Agile solutions provider with expertise in diverse development technologies, such as Angular, the .Net Framework, SQL Server and other cloud friendly data stores. We are certified and have successfully delivered projects across different cloud technology stacks such as Microsoft Azure and AWS, delivering integration and development solutions since 2000.

We work with a number of the UK’s most respected financial institutions to deliver a range of innovative solutions. We have expertise in working with both established businesses as well as start-ups and extreme growth businesses.

Page 1 of 1712345...10...Last »

RECENT POSTS

RECENT JOBS

ADDRESS

509 The Print Rooms
164-180 Union Street
London, SE1 0LH
Phone: 0208 1444 206
Website: synetec.co.uk
Email: info@synetec.co.uk

DISCLAIMER

Important: The information contained in this website is for general information purposes only. Any reliance you place on such information is therefore strictly at your own risk. Synetec Ltd endeavour to keep it up to date and correct.
All images are copyrighted to their respective owners.
Bitnami