Technology posts

What are Artificial Neural Networks in Computer Science

An artificial neural network (NN for short) is a classifier. In supervised machine learning, classification is one of the most prominent problems. The aim is to assort objects into classes (terminology not to be confused with Object Oriented programming). Classification has a broad domain of applications, for example:

  • In image processing we may seek to distinguish images depicting different kinds (classes) of objects (e.g. cars, bikes, buildings etc) or different persons,
  • In natural language processing (NLP) we may seek to classify texts into categories (e.g. distinguish texts that talk about politics, sports, culture etc),
  • In financial transactions processing we may seek to decide if a new transaction is legitimate or fraudulent.

The term “supervised” refers to the fact that the algorithm is previously trained with “tagged” examples for each category (i.e. examples whose classes are made known to the NN) so that it learns to classify new, unseen examples in the future.

In simple terms, a classifier accepts a number of inputs, which are called features and collectively describe an item to be classified (be it a picture, text, transaction or anything else as discussed previously), and outputs the class it believes the item belongs to. For example, in an image recognition task, the features may be the array of pixels and their colors. In an NLP problem, the features are the words in a text. In finance several properties of each transaction such as the daytime, cardholder’s name, the billing and shipping addresses, the amount etc.

It is important to understand that here we assume that there is an underlying real relationship between the characteristics of an item and the class it belongs to. The goal of running a NN is: Given a number of examples, try and come up with a function that resembles this real relationship (Of course, you’ll say: you are geeks, you are better with functions than relationships!) This function is called the predictive model or just the model because it is a practical, simplified version of how items with certain features belong to certain classes in the real world.


Get comfy with using the word “function” as it comes up quite often, it is a useful abstraction for the rest of the conversation (no maths involved). You might be interested to know that a big part of the work that Data Scientists do (the dudes that work on such problems) is to figure out exactly which are the features that better describe the entities of the problem at hand, which is similar to saying which characteristics seem to distinguish items of one class from those of another. This process is called feature selection.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.


Gain momentum in your business with Robotic Process Automation

Outsourcing, in the current scenario, has become an integral part of operations of most major enterprises. The nature of outsourcing, however, is evolving at a fast pace. Organizations must understand these trends and develop strategies to take advantage of these evolutions. They must adapt to the shifting geopolitical policies and economies of labor arbitrage by harnessing emerging technologies and innovations.

Outsourcing is gradually shifting away from the offshoring and labor arbitrage models to more technical and innovative models that optimize productivity and cost-effectiveness while employing the local workforce. According to our recent paper (developed in association with The Outsourcing Institute), automation is becoming the most important technology and currently accounts for more than 10 percent of annual GDP growth worldwide. Robotics Process Automation (RPA) in particular, is quickly becoming an integral part of the new BPO model. As technologies such as predictive analytics, machine learning and cognitive computing are becoming more powerful, RPA is becoming more relevant, reliable and cost-effective.

RPA is expected to dominate the outsourcing industry, as global political and economic forces reduce the viability of geography-based labor arbitrage. Corroborating this hypothesis, our study shows that more than 68 percent of respondents said they would be increasing their spending on RPA.


Outsourcing service providers no longer compete on costs of labor only, but also on building and integrating new technologies in novel ways. Providers who don’t move quickly and innovate will be left behind. As traditional outsourcing models lose their viability, advances in automation are allowing service providers to stay competitive while offering better service to their clients.

A successful deployment of RPA can allow companies to improve productivity, reduce costs, improve governance and increase reliability. However, in order to implement RPA successfully, outsourcers need to have the necessary expertise and planning in place. Creating a robust automation strategy, setting up a Center of Excellence (CoE) with experts, innovating with new functional models and iterating to excellence are keys to continued growth in the coming years.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.



Procurement needs to know about robotic process automation

Robotic Process Automation (RPA) vendors emphasise their product’s capacity to replace human operators, using phrases like “digital workforce.” In simple terms, RPA is a software application that runs on an end user’s computer, laptop or other device, emulating tasks executed by human operators.

Its purpose is to integrate or automate the execution of repetitive, rule-based tasks or activities. RPA does not require development of code, nor does it necessitate direct access to the code or database of the applications.

Current Robotic Process Automation Use

Most current RPA implementations are in industry-specific processes such as claims processing in insurance, and risk management in financial services. These processes, and their associated tasks, are usually high-volume, structured, repetitive and implemented on old technology.

Normally, the processes are extremely stable. There is no technology migration or modernisation roadmap involved, and IT-led integration would be difficult and expensive.

At present, the leading non-industry-specific RPA application is the financial close and consolidation process. According to our purchase-to-pay research, 23 per cent of companies are at the earliest stages of adoption, i.e., either in a pilot or with the technology partially rolled out.

The Best Processes for RPA

It is not the type of business process that makes for a good candidate for RPA, but rather the characteristics of the process, such as the need for data extraction, enrichment and validation.

Activities requiring integration of multiple screens, as well as self-service inquiry resolution, are also ripe for RPA. The key is that RPA is best deployed in a stable environment where no changes to the systems are on the horizon.

Other possible choices include processes requiring multiple software applications to execute different, but repeatable, activities and tasks.

RPA Pricing Trends

The pricing model for RPA is still evolving. Today, vendors are pricing RPA based on the cost of the full time equivalent (FTE) staff member it is replacing. For example, an RPA vendor may quote a price per robot that is one-third the cost of an offshore resource doing the work.

Onshore FTE pricing is being quoted closer to one-ninth, or 11 per cent, of the cost. This pricing model, developed to compare the cost of outsourcing a process versus automating it with RPA, essentially positions Robotic Process Automation as a service, not a software solution.

This model is inconsistent with industry standards governing the way software is typically priced. Therefore, we encourage buyers to seek an alternative gain sharing model where possible. This will both mitigate the risks of early adoption, and provide a strong incentive to the supplier to deliver results.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.


Three Main stages to Robotic Process Automation success

Robotic Process Automation offers a variety of ways to increase the degree of automation. For organizations, however, robotics is a new phenomenon and its implementation may arouse suspicion and resistance to change. Therefore, business management should quickly increase knowledge about RPA among management and employees in order to ensure that the potential benefit is realizable.

RPA software, just like any other business software, is used and steered by people. While the degree of automation is being increased in, say, payroll computation and financial administration processes through RPA, robotics should shift towards people’s work and be integrated into it. RPA is based on well-defined processes and logical rules related to them. These allow the robot to work either a) automatically in the background or b) initiated by a human. When changes are made to a process or to the business software used in the process, the robot has to be retrained for its task as well. This training will be a new special skill that companies must be able to obtain.

Three stages on the path of progress in Robotic Process Automation

We have defined three stages for the path of progress in robotics. Each of these contains different requirements for the organization’s ability to develop and learn.

Stage 1

In the first stage, the organization sets out to test and pilot RPA in its processes. In this stage, it is usually a good idea to use a consultant and an expert with special know-how in the field of RPA as well as a vision about how Robotic Process Automation could benefit your company. It is typical to select one function and a few processes as starting points for piloting. With the help of the consultant, the benefits of RPA can be estimated quickly and an action plan can be prepared for the development of RPA automation.

The key is to increase your organization’s understanding of RPA and gather experiences.

Stage 2

In the second stage, the organization expands the utilization of RPA to several functions and processes. The consultant can help to define complex sets of rules and assume part of the workload in training the robots. In this stage, the key is to share the information and know-how obtained in the pilot with the organization and to increase internal RPA capabilities in a determined manner.

Stage 3

In the third stage, RPA has become the standard procedure for increasing the organization’s productivity, and it is managed through a centralized center of excellence or function-specific RPA teams. The organization possesses plenty of internal intellectual and knowledge capital on robotics and is able to maintain and develop these self-sufficiently. The key is to standardize RPA and to measure and manage competence and capabilities.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.


Improving your Business Functions with Robotic Process Automation

Everyone thinks that Robotic Processes and Artificial intelligence are something from the future, although we all know that robotic technology have been successfully used in manufacturing for many years, building our automobiles.  But, the question is can robotic process technologies be used in other industries?

Deployment of robotic process automation industry wide

Robotic process automation (RPA), and it has been around for decades.  The technology has emerged recently as a significant trend among many industries. RPA is even more prevalent now due to more affordable robotic systems. According to industry experts, robotic systems used to be very expensive and could only be used in high-value products to recover the costs of their use. More reasons to move forward with Robotic processes is a simple fact that there is an increasing lack of low-cost human labor.

RPA is the method of employing robotic systems to “perform tasks repeatedly and usually [at] high speed, without human intervention.” This technology can take over high-volume, repeatable and often tedious tasks that were previously performed by humans. People are then freed up to do higher-value work which require empathy and a human touch.

Many industries require customized products to suit their needs, this is also true with RPA. With advances in robotic systems, cameras, laser systems and AI driving down costs across business ecosystem, businesses can produce many types of items, but at low quantities. Deploying such intelligent robotic systems are easier and less costly now than in the recent past.

Creating a capable virtual workforce with your software apps

Besides hardware, RPA can be applied to software as well. Companies can create a virtual workforce that is rules-based with heuristic algorithms. Human resources and administrative processes can be made smoother and more efficient as a result.

The financial industry is not the only industry that can learn from the robotic process automation. Banks have been using the robotic process automation to automate anything from data entry, report generation, and credit card processing to fraud detection and audit support, due to this integration, they have increased efficiencies and reduced costs. Deutsche Bank is using RPA to remove manual processes from the back and middle office processes.  They are focused on using this technology to increase efficiencies, and remove the need for tedious tasks performed by people.

And in the insurance industry, which typically requires a lot of paperwork, RPA can assist in auto enrollment, product administration, and policy document data transfers.

While RPA automates manual tasks, humans are more adapt in roles that require emotion and empathy; healthcare and medicine come to mind and may not be the industry to take advantage of RPA.   There will always be industries which will benefit from robotic process automation.

Like other forms of automation, RPA can augment existing workforces, and essentially allow them to do more with less. RPA will not replace humans, but will help reduce errors and operate for longer periods of time. This increase in productivity won’t likely be realized without the integration of RPA.  This increased productivity and lower costs will have a positive effect throughout the business process.

As technologies like RPA advance, it seems the robot uprising will probably be more behind-the-scenes than anyone expects.   RPA gives a competitive edge, unlocking higher business quality, productivity, and agility.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.


Robotic Process Automation (RPA) and Enhanced Customer Experience

Traditionally processing of loans requires the customer to fill up many forms and submit required sets of documents. These would then be sent for processing, where someone would re-enter the details (with possible errors), then back-office managers will do checks (credit scores, property details, etc.) and send the application for approvals. Some of these tasks may be outsourced to 3rd party vendors, which might lack proper control/ audit procedures. The overall process takes a few weeks, after which the customer gets a feedback on the status of his loan application.

When a robotic process automation software is implemented, the bot can take over the complete process – from uploading the scanned documents, verifying e-signatures, verification/checks and scores for automatic approval or rejection recommendations with complete audit trail.

RPA reduces the turn-around time to a few days (conditional approvals can be provided on quick screening) from weeks!

Customers can track their application stages and status online and are informed of process updates and the stages their application is flowing through. This not only creates a beneficial customer experience, but allows the Mortgage provider to invest these savings in front line / customer facing resources – which along with the robotic process automation software will drive up revenue as well as customer experience.

Metrics and continuous improvement

Organizations can set-up metrics to measure customer experience (number of complaints, changes, rework, etc) and work on a continuous improvement plan for customer experience journey. A comparative metrics between pre and post-RPA implementation will clearly bear out benefits of investment in RPA and help in understanding how investments in direct customer touchpoints will further improve customer experience.

Business process automation is not an end in itself to improvements in customer experience. A continuous improvement plan with improvements and maintenance of BOTS will further improve customer satisfaction. The BOTS need to be maintained and quality checked time to time, as every automated systems will encounter “exceptions”, due to change in processes or new data sets being introduced.

We at Veltrod, help our customers implement the robotic process automation software, provide BOT support and maintenance. We help our customers enhance their end user experience and work on a continuous improvement plan to target customer delight.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.


Why automate business processes with Robotic Process Automation (RPA)

RPA automates high volume, repetitive manual (clerical) processes exactly as performed by human workers. Work is performed by software robots or “bots” that carry out processes between applications, portals, websites, and other data sources. Essentially, software robots move data around between applications just as human workers do, clicking from window to window or working between machines or monitors.  

How does RPA software work?
RPA uses presentation layer automation software—a technology that mimics the steps of a rules-based, process to create a software robot. Software robots collect, aggregate, extract and input information from all your data sources exactly as a human would. Because RPA automates by mimicking human actions at the user interface level, RPA eliminates having to develop new APIs or replace existing applications. 

Processes are built by showing the robots what to do step-by-step in exactly the same way a human would use the end user systems—no coding required. RPA software is used to train the robot to read the various screens it needs to work with. For example, one software robot can be trained to gather all the relevant files to document a loan, combine them into a PDF, and then send an email to notify the appropriate person that the file is ready. Once it is deployed, the robot repeatedly performs the defined task as any human would, only much faster and with less chance for error.

What types of work can be automated with RPA?
By definition it requires logging in and out of different systems to complete the process (or even a single task) and often involves third party systems or otherwise environments which cannot be integrated through a programmatic interface. Instead people do it, with swivel chairs and sticky notes, and as a result the design of the related rules and workflows are based on how the applications were built rather than the actual objectives of the end-to-end process which span across them.”hqdefault

RPA is best suited for work with repetitive steps that do not require any meaningful analysis and where users are shifting back and forth between different application interfaces or multiple systems from different machines as part of the task or process. RPA is particularly suitable to high-volume, highly transactional processes typically found in back-office tasks (such as finance, procurement, supply chain, accounting, customer service and human resources). 

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.


Benefits of Robotics Process Automation

Reduces Reliance on IT

In addition to the obvious savings in terms of operational overheads, IT reliance can and should be reduced by an RPA program. One of the key tenets for RPA is that it empowers the business in managing their operation and removes what at times are viewed as the dependency on, and constraints imposed by, IT.


Another major benefit of RPA over labor arbitrage is the ability to scale quickly, both in terms of expansion and contraction. The on boarding of new staff members is both costly and time-consuming. If this cost is considered in light of some of the high turnover within BPO markets, then it becomes a real drag on an organization’s ability to get to its optimal operation cost. There are no on boarding costs for a robot. To be fair, though, there are project costs for implementing the robotic processes. The volume of business throughput is no longer dependent on human resource constraints with RPA.

Consistency of Process

RPA also offers the benefit of consistency of process. Even the best scripted, highest educated call center staff can still interpret a process in different ways. Depending on which staff member the customer speaks with on a given day, that customer can have a fundamentally different experience. RPA homogenizes the experience. This is particularly true in areas of high impact, such as risk assessment.

RPA introduces repeatable processes that are subject to rules and preciseness that even the smartest individuals cannot maintain. The renewal process of an income protection claim requires several steps that should be identical for each customer (identification of the customer, confirmation of disability, validation of product rules, etc.). These steps can all be automated to ensure that the customer, instead of having multiple points of contact with the insurer, has only to initiate the renewal request. RPA takes care of the rest.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.


How Natural Language Processing is Changing Business Intelligence

From onsite customer behavior, to seasonal or even daily trends, the typical data warehouse can contain an eclectic mix of information. The insights gain from these data have propelled businesses into a new realm of customer understanding, but limiting analytics to this type of highly structured format excludes the majority of the data that’s being created today.

80 percent the data being created is unstructured. It’s generated from conversations between customer service reps and on social media, among other places. To mimic the advantages gained from harnessing transactional data, organizations are turning to natural language processing (NLP) technology to derive understanding from the myriad of unstructured data available online and in call-logs.

As early as 1960, engineers were working to design programs that could derive meaning from language. By the 1980s, natural language processing had grown enough to harness some meaning from conversation, but only in the form of rigid IF-Then rules. This format was incredibly time consuming to write, not to mention limited in its scope.

Driven by highly structured languages, it’s always been difficult for machines to grasp the context of human speech. But machine learning has helped computers parse the ambiguity of human language. With the advent of advanced statistical algorithms, programs are now capable of using statistical inference to make predictions on what was meant in conversation by calculating the probability of certain outcomes. And the brilliance of inference and machine learning is that the program can continually improve itself the more data it processes.
1_t1EFFcUdFIW8CaTVLTsCtQWhat does this mean for business? It signifies that all the insights hidden in unstructured data are becoming more attainable with each technological advance. It means that qualitative data is now quantifiable.

How Businesses Use NLP

A subset of natural language processing, natural language understanding is concerned with the reading comprehension of machines. By using the aforementioned statistical inference model, software developers are helping make natural language understanding a reality.

The most common application of natural language understanding is text analysis, also known as sentiment analysis. While transactional data helps organizations predict what actions customers will take, it fails to offer much insight into how they felt during the process, leaving significant gaps in understanding the customer relationship. That’s why businesses are most concerned with comprehending how their customers feel, not just how they’re going to act.

Sentiment analysis can most commonly be put to work gathering insight from social media. With millions – in some cases even billions – of current and potential customers online, there’s tons of data being created each day that brands can harness. Basic sentiment analysis tools like Digimind can search the web for mentions of your brand and quantify whether the context was positive, neutral, or negative. Digimind digs deeper into context by ranking the importance of the source based on their social media clout.

Email filters are another common application of NLP. By analyzing the emails that flow through their servers, email providers can calculate the likelihood that an email is spam based its content. Call centers are another area rich with unstructured data. Whenever customer representatives engage callers, those callers list specific complaints and problems. Mining this data for sentiment can lead to incredibly actionable intelligence that can be applied to product placement, messaging, design, or a range of other use cases.

Natural language processing and sentiment analysis has even found its way into higher education. 

A text analysis tool set up to work with Excel, the Dean was able to isolate key noun phrases from his students’ blogs and aggregate them in Excel for Semantria to analyze. With the aid of Semantria, Seattle Pacific’s Associate Dean of Education was able to connect specific noun phrases with both quality reflection and classroom performance. Specifically, he found a greater range of phrases was correlated with better classroom performance. Now, Seattle Pacific encourages their School of Education candidates to write more elaborate posts, thereby encouraging more in-depth analysis of their teaching habits.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.


How Natural Language Processing Has Changed SEO

Have you ever wondered how Google, Facebook and Pinterest seem to get you the results you’re looking for even when you don’t type the words exactly right? If you have, you have probably inadvertently seen natural language processing at work.

During the last few years, natural language processing has come to change the way the web works in a fundamental way: shifting discovery of content from an explicit keyword based search to discovery based on context and intent. This is life changing for the SEO professional.

A brief history of natural language processing

Natural language processing, or NLP, is a concept that had its genesis in philosophy and matured in the realm of linguistics. For decades, NLP did not make its way into mass market applications because of technical limitations. One of these was the need for “supervised learning,” which is humans teaching computers how to resolve conflicts of understanding. The last two decades have seen significant advances in the field of computer science. These advances include the development of semantic web technology and machine learning. Together, they have made NLP a reality for many more uses, like the applications we rely on with our mobile phones.

How does NLP impact web searches and SEO trends?

Because it’s the first search engine that most people think about, let’s begin with Google search.

Google has at least two ranking algorithms in place that run in parallel. The first is the traditional PageRank algorithm. The second algorithm is often referred to as the “mobile” algorithm. It covers many new factors, including the context of your query as well as the search terms.

Most SEOs worth their salt know that the original PageRank algorithm has driven many SEO strategies that include on-page optimizations, like page titles and the pursuit of backlinks. These are still effective mechanisms for surfacing content, particularly for long tail topics.

Google’s newer algorithm looks to determine your search intent from several factors. These include:

  • What type of device you are using (desktop, mobile phone, tablet)
  • Are you typing or using your voice
  • Number of consecutive searches
  • What else people are searching for

Research shows that topics are more important than keywords

What about everything you’ve been hearing about it being time to start looking at topics instead of keywords? Topic based approaches take into account the fact that we all think differently and might not be using the same words to find things when searching the internet. They also tend to support searches by less sophisticated users.

Tips for making content marketing more effective for natural language search

There is not going to be a single recommendation for every business to follow for optimizing their site for the new world of natural language search. To keep things simple, I would recommend the following tips when crafting content with good SEO in mind:

  • Plan content around topics and vary the use of related terms rather than focusing on one single keyword.
  • Design for mobile consumption.
  • Deliver the most useful format for your visitor.
  • Liberally link to other authoritative and related resources.

Veltrod is specialized in providing deep learning solutions across the domain. Write to for free consultation and quotes for business needs.

© Copyright 2013 Veltrod Scroll Top