Adopt the Right Testing Strategies for AI/ML Applications

The adoption of systems based on Artificial Intelligence (AI) and Machine Learning (ML) has seen an exponential rise in the past few years and is expected to continue to do so. As per the forecast by Markets and Markets, the global AI market size will grow from USD 58.3 billion in 2021 to USD 309.6 billion by 2026, at a CAGR of 39.7% during the aforementioned forecast period. In a recent Algorithmia Survey, 71% of respondents mentioned an increase in budgets for AI/ML initiatives. Some organizations are even looking at doubling their investments in these areas. With the sporadic growth in these applications, the QA practices and testing strategies for AI/ML applications models also need to keep pace.

An ML model life-cycle involves multiple steps. The first is training the model based on a set of feature sets. The second involves deploying the model, assessing model performance, and modifying the model constantly to make more accurate predictions. This is different from the traditional applications, where the model’s outcome is not necessarily an accurate number but can be right depending on the feature sets used for its training. The ML engine is built on certain predictive outcomes from datasets and focuses on constant refining based on real-life data. Further, since it’s impossible to get all possible data for a model, using a small percentage of data to generalize results for the larger picture is paramount.

Since ML systems have their architecture steeped in constant change, traditional QA techniques need to be replaced with those focusing on taking the following nuances into the picture.

The QA approach in ML

Traditional QA approaches require a subject matter expert to understand possible use case scenarios and outcomes. These instances across modules and applications are documented in the real world, which makes it easier for test case creation. Here the emphasis is more on understanding the functionality and behavior of the application under test. Further, automated tools that draw from databases enable the rapid creation of test cases with synthesized data. In a Machine Learning (ML) world, the focus is mainly on the decision made by the model and understanding the various scenarios/data that could have led to that decision. This calls for an in-depth understanding of the possible outcomes that lead to a conclusion and knowledge of data science.

Secondly, the data that is available for creating a Machine Learning model is a subset of the real-world data. Hence, there is a need for the model to be re-engineered consistently through real data. A rigor of manual follow-up is necessary once the model is deployed in order to enhance the model’s prediction capabilities continuously. This also helps to overcome trust issues within the model as the decision would have been taken through human intervention in real life. QA focus needs to be more in this direction so that the model is closer to real-world accuracy.

Finally, business acceptance testing in a traditional QA approach involves the creation of an executable module and being tested in production. This traditional QA approach is more predictable as the same set of scenarios continue to be tested until a new addition is made to the application. However, the scenario is different with ML engines. Business acceptance testing, in such cases, should be seen as an integral part of refining the model to improve its accuracy, using real-world usage of the model. 

The different phases of QA

Three phases characterize every machine learning model creation:

The QA focus, be it functional or non-functional, is applied to the ML engine across these 3 phases.

  • Data pipeline: The quality of input data sets has a significant role in the ability to predict a Machine Learning system. The success of an ML model lies in the testing data pipelines which ensure clean and accurate data availability through big data and analytics techniques.
  • Model building: Measuring the effectiveness of a model is very different from traditional techniques. Out of a specified number of datasets available, 70-80% is used in training the model, while the remaining is used in validating & testing the model. Therefore, the accuracy of the model is based on the accuracy shown on the smaller of datasets. Ensuring that the data sets used for validating & testing the model are representative of the real-world scenario is essential. It shouldn’t come to pass that the model, when pushed into production, will fail for a particular category that has not been represented either in the training or the testing data sets. There is a strong need to ensure equitable distribution and representation in the data sets.
  • Deployment: Since the all-round coverage of scenarios determines the accuracy of an ML model and the ability to do that in real life is limited, the system cannot be expected to be performance-ready in one go. A host of tests need to be done to the system like candidate testing; A/B testing to ensure that the system is working correctly and can ease into a real-life environment. The concept of a sweat drift becomes valid here whereby we arrive at a measure of time by when the model starts behaving reliably. During this time, the QA person needs to manage data samples and validate model behavior appropriately. The tool landscape that supports this phase is still in an evolving stage.

The QA approaches need to emphasize the following for ensuring the development and deployment of a robust ML engine.

Fairness:

The ideal ML model should be nonjudgmental and fair. Since it depends largely on learning based on data received from real-life scenarios, there is a strong chance that the model will be biased if it gets data from a particular category/feature set. For example, if a chatbot with learning ability through ML engine is made live and receives many inputs that are racist, the datasets that are being received for learning by ML engine are heavily skewed towards racism. The feedback loops that power many of these models ensure that racist bias comes into the ML engine. There have been instances of such chatbots being pulled down after noticeable differences in their behavior.

In a financial context, the same can be extended to biases being developed by the model receiving too many loan approval requests from a particular category of requestors, as an example. Adequate efforts need to be made to remove these biases while aggregating or slicing and dicing these datasets and adding them to the ML engine.

One approach that’s commonly followed to remove the bias that can creep into a model is by building another model (an adversary) that understands the potential of bias from the list of various parameters and incorporates that bias within itself. By frequently moving back and forth between these two models with the availability of real-life data, the possibility of a model that removes the bias becomes higher.

Security:

Many ML models are finding widespread adoption across industries and are already beginning to be used in critical real-life situations. The ML model development is very different from that adopted for software development. It is more error-prone on account of loopholes that can cause malleable attacks and a higher propensity to err on the wrong side on account of erroneous input data.

Many of these models do not start from scratch. They are built atop pre-existing models through transfer and learning methods. If created by a malicious actor, these transfer learning models have every possible way of corrupting the purpose of the model. Further, even after the model goes into production, malicious intent data being fed into the model can change the prediction generated by the model.

In conclusion, assuring the quality of AL/ML-based models and engines needs a fundamentally different approach from traditional testing. It needs to be continuously changing to focus on the data being fed into the system and on which predictive outcomes are made. Continuous testing, focusing on the quality of data, the ability to affect the predictive outcome, and remove biases in prediction is the answer.

(This blog was originally published in EuroStar)

Quick Wins in Enterprise Digital Transformation (yet often ignored) – Intelligent Automation

The modern workplace is seeing widespread usage of machines and automation. Enterprise digital transformation, Artificial intelligence (AI), and automation are changing the tide for businesses globally. This means a significant change in the work culture as employees will have to acquire new skills and adapt to the advanced capabilities of machines. 

As per a recent study1 involving over 600 business leaders from 13 countries, more than 50 percent of respondents confessed to having already invested over $10 Million in intelligent automation projects. The AI market globally is presently growing at a CAGR of 40%, all set to touch $26.4 Billion by 2023.

AI,  along with robotic process automation (RPA), voice recognition, natural language processing (NLP), and machine learning (ML), is allowing businesses to blend automation with human capabilities successfully to create intelligent working environments. 

Automation is driving agility for businesses giving them the much-needed competitive edge over others with quick decision-making powers. Clearly, decision velocity powered by AI-driven insights gives you data supremacy to lead in a highly volatile market.

Making a case for Intelligent Process Automation (IPA)

When automation meets artificial intelligence, you get intelligent process automation to scale up your business. While it allows you to off-load routine, repetitive tasks, it empowers better guardrails for all your automation initiatives. It takes the uncertainty out of the picture and enables more personalized execution and processes.

Intelligent automation enhances the overall customer experience. The speed of response has often been a critical consideration while evaluating the customer experience. Intelligent automation is helping organizations meet customer expectations with personalization. Through customized offers, services, and content, businesses are acquiring and retaining customers.

What do the right IPA endeavors ensure?

  • Agile services due to a significant reduction in processing time
  • Greater flexibility and scalability for being able to operate round the clock with capabilities to scale up and down as required
  • Improved quality control due to greater traceability of events and instances and checks at different levels
  • Increased savings and productivity due to a high level of automation
  • Clear, actionable insights to predict and improve drivers of performance

While there is unanimous agreement on the benefits of intelligent automation, not everyone has leveraged these benefits across the organization. What you need is an enterprise-wide approach that promotes a new way of working.

Adding intelligence to the digital mix

A highly automated world does not focus on reducing the headcount but increasing its potential to do more in an agile manner to solve the business challenges of tomorrow. It relies on structured and unstructured data the company collects from the public domain and other stakeholders rather than depending on traditional methods.

Intelligent automation compels you to rethink key business processes. The sales and marketing team gets deeper segmentation to target and sell through advanced analytics. Those working to strengthen the supply chain get to improve production and distribution by leveraging technologies like cloud and analytics across the value chain. Planning and development teams, on the other hand, rely on data-driven insights to integrate them into product performance and boost innovation.

Alibaba Group2 is a classic example of what you can achieve with intelligent automation.

After making significant strides in eCommerce and retail, it has further revolutionized its business processes with its ‘Industrial Vision AI‘ solution for manufacturing and production. It allows the company to inspect raw materials thoroughly to detect minute defects, resulting in a 5X increase in production efficiency. Its automated warehouse is managed entirely by robots taking precision and efficiency to a whole new level.

Regardless of your goal, you need to create a strategic roadmap to align it with your business priorities. This is not possible unless you assess your digital maturity.

What is the role of IT in successful IPA transformation?

Intelligent process automation (IPA) is a melting pot of technologies enabling significant gains for businesses worldwide. IPA should not be confused with robotic process automation (RPA) as unlike RPA that performs repetitive, automated tasks based on predefined rules and inputs, IPA can understand the context, learn, and iterate to support informed decision-making using unstructured and structured data.

Those who have been able to get the full value of IPA have been the ones who have put IT leaders at the helm of their IPA endeavors. CIOs need to strengthen their core with IPA programs to support automation.

Here’s what we recommend:

  1. Assess the high-level value potential

You may start with help-desk requests since that’s where a significant amount of incidents originate. While tickets with low difficulty levels are resolved immediately, those with more complexity are often escalated to specialized teams. Determine how many such requests were handled the previous year, and by multiplying them by the average handling time (AHT) required, you can evaluate the value of this whole exercise.

For instance, an organization with a significant number of requests for password reset or access can leverage RPA bots that work across multiple applications via the user interface to automate ticket resolution and free up employee capacity. Reducing resolution times and a drop in costs associated with outsourcing help-desk support will thus improve performance and profits.

The effort required for these activities often varies. Everything needs to be evaluated critically from backups and patching to security audits and upgrades to understand the effort involved and the value you can garner by planning activities for automation.

  1. Identify the use cases best suited for IPA

Let’s consider the same example mentioned above. In order to automate incidents, organizations need to first identify the ones ideal for automation. An organization may be effectively logging incidents in detail, but due to the large numbers and complexities, support teams may not respond quickly and effectively.

AI can make sense of the chaos and understand the reasons behind the alerts. It may be trained to make appropriate recommendations or even make better decisions to ensure suitable responses.

  1. Elevate customer experiences with better service

AI and automation are changing the customer service landscape for every industry, from retail to aviation. Boeing has a fleet of passenger service robots that operate via sensors installed in their bodies. They are doing their best to reduce the manual work of cabin crews. Though experts argue a human perspective is required for these robots to do what humans can.

The key is to understand the power of automation and integrate it seamlessly into processes and workflows to complement human efforts and endeavors perfectly, as we did in the case of one of our clients Surge Transportation.

The company links shippers and carriers and has an automated tracking and monitoring system to assign loads. But the pricing and quotation were being done manually. This drained their resources, led to a huge turnaround time, and left a long log of emails, calls, and paper trails.

Trigent critically evaluated the complexities in its pricing mechanism to bring down the turnaround time to less than a second. Apart from 100% pricing accuracy, the company improved profits by 25%, revenue by 40%, and reduced the load processing time by 91%. With seamless carrier integration, the company now processes 4000 more loads per day.

Other use cases where AI and Automation are driving value

Cashier-less stores

Amazon is popularizing the concept of cashier-less stores with Amazon Go and Just Walk Out. Robotization of stores helps save operational expenses and gives shoppers a smart shopping experience.

Automated medical appointment scheduling

No-shows have been the cause of losses of over $150 billion a year for the U.S. healthcare system with every unused time slot costing individual physicians $200 on an average. No-shows also impact the health of patients since continuity of care is interrupted. IPA challenges traditional scheduling methods by ensuring error-free appointment scheduling based on the nature of the illness, the convenience of patients, and the availability of doctors and healthcare facilities. While patients get to choose a date and time for different health issues, follow-up appointments can be scheduled automatically along with reminders.

Automated supply chains

The ideal supply chain is where there is neither wastage nor out-of-stock scenarios. In tandem with machine learning, AI predicts demand based on location, weather, trends, promotions, and other factors. Revenue losses of up to $4Trillion have been caused due to supply chain disruptions following the pandemic with 33% attributed to commodity pricing fluctuations as per a report.

The automobile giant Toyota is using AI in its manufacturing environment to address waste control with its ability to predict when excess parts, products, and practices threaten to impede work.

Intelligent Automation is clearly the Winner streak!

The potential value of AI and automation is immense for different sectors and will vary depending on the type of industry, availability of abundant and complex data, use cases, and other factors. To get the most out of your automation initiatives, it is however important to tide over organizational challenges with the right mindset and approach.

Create impact and value with Trigent

Trigent with its team of technology experts empowers you to stay relevant and competitive. It is equipped with insights and intelligent solutions to dramatically boost your bottom line and improve customer engagement.

Allow us to help you grow your business and increase revenue with strategies and solutions that are perfect for you.

Call us today for a business consultation

References
1. https://www.analyticsinsight.net/intelligent-automation-accelerating-speed-and-accuracy-in-business-operations/
2. https://datacentremagazine.com/technology-and-ai/alibaba-group-adopts-ai-and-automation-singles-day

AI Implementation Checklist – 5 Points to Evaluate Before Adopting AI in Your Organization

Artificial intelligence is now all around us in household gadgets as well as business workflows. AI adoption is rampant across sectors; the global artificial intelligence market is expected to reach $ 266.92 billion by 20271 at a CAGR of 33.2% during 2019-2027. Nearly half of the respondents who had participated in a survey confessed to being interested in AI implementation and machine learning to improve data quality.

No doubt, artificial intelligence is the proverbial genie that does everything we want it to do without even rubbing the magic lamp. But the lack of nuance and failure to spell out caveats can result in AI systems that will make us think twice before we wish for anything.

Believe it or not, misaligned AI can be a nightmare.

A classic case is YouTube2, with its AI-based content recommendation algorithms that led to users accusing it of radicalization. Its constant upping-the-ante approach led users to extreme content in a bid to maximize viewing time. So videos on vegetarianism led to veganism, and jogging searches resulted in ultramarathons. This unintentional polarizing and radicalizing highlights one significant challenge: we have yet to define the goals accurately for our AI systems!

The sad truth is that we don’t even know what we want, at least not from our autonomous systems and gadgets and other possessions. For instance, a self-driving car may be too slow and brake too often just the way it was designed to prevent itself from colliding with nearby objects. But the object could be as insignificant as a paper bag that was blown away by the wind.

What we need is goal-oriented AI born with a solid sense of purpose with excellent human-machine coordination. But only after you have answered the question- Do I really need AI?

Here’s is your ultimate AI implementation checklist

AI has ample scope in many sectors. AI can interact on your behalf with customers, as in the case of chatbots, or help healthcare providers diagnose cancer and other ailments. If leveraged well, it can help you turn a new leaf in critical interactions with your customers. Understanding the potential of AI and applying it to enhance critical business values can make a world of difference to your business. The key is to know where you stand and whether AI can help you attain your business goals.

Identify the purpose

Organizations with successful AI implementations are usually the ones that have assessed its financial impact or conducted a thorough risk analysis for AI projects. Having the right metrics in place gives you a sneak peek into the risks and benefits of AI implementation and how it would perform in those chosen areas. While it may not guarantee a positive ROI, it gives you a fair idea about what to expect. 

Accuracy, for instance, is an important metric, but it’s not enough to understand how well your AI systems are performing. You need to correlate  AI metrics to business outcomes to ensure you get the most out of your investments. 

The smart pricing tool created by Airbnb to eliminate pricing disparities between black hosts and white hosts presents a classic example. While the AI-based system performed the assigned tasks with precision, the business results fell short – widening the gap further by 20%. 

Appoint mixed-role teams for all AI initiatives

Those who have implemented AI successfully will tell you how crucial it is to build mixed-role teams comprising project managers, strategists, application designers, AI researchers, and data scientists to ensure a diversity of thought and skillsets. As per a Garnet Research Circle survey3, skills are the first barrier to AI adoption, and 56 percent of respondents believed new skills are required for new and existing jobs.

AI needs experts for it to evolve to its best version. TayTweets, a promising chatbot by Microsoft, was nothing but fun, and people loved talking to her. Until, of course, she became the nastiest chatbot ever in less than 24 hours, responding with offensive tweets. It demonstrates how horribly things can go wrong when AI and ML go awry when left unchecked.

Diversity in technical acumen enhances the value of AI to customers since the people working with AI know-how and where it should be used to have the most significant impact. Whether you want to hire new people or train existing ones for newer roles and responsibilities is something you will have to decide based on the business initiatives you have in mind.

Make a business case for AI

Businesses need AI for different reasons ranging from data security and fraud detection to supply chain management and customer support. You need to identify the use cases and applications to determine how AI can be effectively used. Organizations depend on AI to analyze contextual interaction data in real-time and compare it with historical data to get insights and recommendations.

Data plays a pivotal role in every aspect of a business. While a lot of emphases is placed on coding, math, and algorithms, many organizations are not able to apply the data acquired effectively in a business context. You will have to understand who you are building these solutions for and what technology framework you will require to do so.

As Moutusi Sau, principal research analyst at Gartner4, points out, “Business cases for AI projects are complex to develop as the costs and benefits are harder to predict than for most other IT projects. Challenges particular to AI projects include additional layers of complexity, opaqueness, and unpredictability that just aren’t found in other standard technology.”

Assess your AI maturity

It is impossible to arrive at a strategy without evaluating where you stand against the AI maturity model. Once you know it, you can decide the next steps. Typically, the AI maturity model has five levels:

Ø Level 1 – There is awareness in the organization, and AI implementation is being considered, but no concrete steps have been taken in that direction.

Ø Level 2 – AI is actively present in proofs of concept and pilot projects.

Ø Level 3 – AI is operational, and at least one AI project has made its way to production with a dedicated team and budget. 

Ø Level 4 – AI is part of new digital projects, and AI-powered applications are now an essential part of the business ecosystem.

Ø Level 5 – This should be the ultimate goal where AI is now ingrained in the organizational DNA and plays a transformative role for your business. 

Look beyond the hype

AI can cause ‘cultural anxiety’ as a significant shift in thought and behavior is necessary for successful AI adoption. A compelling story to help employees understand how AI would be beneficial to all is necessary to ease the resistance they might feel towards the change.  CIOs should recognize their fears and anxiety of the possibility of being replaced by machines and encourage an open dialogue with team members. This will build trust and help determine if the organization is ready for AI.

The hype around AI itself can sometimes be the biggest problem as organizations hurry to hop onto the AI bandwagon with insufficient understanding of its impact. Explains Whit Andrews, distinguished vice president analyst at Gartner, “AI projects face unique obstacles due to their scope and popularity, misperceptions about their value, the nature of the data they touch, and cultural concerns. To surmount these hurdles, CIOs should set realistic expectations, identify suitable use cases and create new organizational structures.” 

 AI to Impact

The biggest mistake organizations make when they invest in AI is that they have too many expectations and little understanding of AI capabilities. Rather than getting caught in the hype, you have to be realistic and evaluate its role critically in furthering your business objectives.

AI is an expensive investment that will give you good returns if you know how to use it. A lot of tools are good, but not every AI tool is suitable for your business. What you need is the right AI implementation strategy created with professional help from those who know AI like the back of their hand.

Adopt AI with Trigent

Artificial intelligence is a defining technology that can be successfully integrated into business workflows and applications. We at Trigent have been helping organizations from diverse sectors, including healthcare, retail, BFSI, and logistics, create AI operating models that are optimized for faster and effective outcomes. 

We can help you, too, with everything from strategy and implementation to maintenance and support.

Call us today to book a business consultation

References

  1. https://www.fortunebusinessinsights.com/industry-reports/artificial-intelligence-market-100114
  2. https://firstmonday.org/ojs/index.php/fm/article/view/10419/9404
  3. https://www.gartner.com/smarterwithgartner/3-barriers-to-ai-adoption/
  4. https://www.gartner.com/smarterwithgartner/how-to-build-a-business-case-for-artificial-intelligence/

To Opt or Not? Can Traditional Industries Use Machine Learning to Garner Business Insights?

Machine learning is a scientific discipline that uses algorithms to learn from data instead of relying on rules-based programming. It works in three stages, i.e. data processing, model building & deployment, and monitoring, with machine learning binding the three together. The power of machine or deep learning cannot be underestimated and as Alexander Linden, Research Vice President of Gartner says, ‘Deep learning can give promising results when interpreting medical images in order to diagnose cancer early. It can also help improve the sight of visually impaired people, control self-driving vehicles, or recognize and understand a specific person’s speech’.

To Opt or Not

Traditional industries have many processes which are governed by rules-based software. This approach is limited in its ability to tackle complex processes. If the rules-based learning can be substituted with self-learning algorithms, then valuable patterns and solutions would emerge.

As a result of digital data and Internet of Things there is a proliferation of data. If you believe this data will help you make intelligent decisions based on patterns, add machine learning. There is no need to add it otherwise as it can make an existing business complicated. Starting with the smaller pieces of the puzzle is better than jumping into it head on. For example, one can collate information from regular reports, apply machine learning to forward-looking predictions.

Machine learning can be useful to detect anomalies, enhance customer services and recommend new products. Manufacturing companies, for example, can benefit from machine learning by self-examining videos where defects can be spotted and automatically rerouted.

Recent developments in machine learning suggest a future in which robots, machines, and devices will be able to operate more independently if they run on self-learning algorithms. This would have far reaching effect in terms of improved efficiency, and cost savings.

Related: Reshaping your business with AI

Machine learning works best on specific tasks where input and output can be clearly stated. If an organization has a sufficient amount of data, with enough variation, machine learning can produce meaningful approximations.

Finally, it is the technical barriers that become the biggest hurdle in the transition process. To address the actual challenges and the perceived ones, companies need to identify expert data analysts who are capable of developing the intricate algorithms that machine learning requires. It will also require a team of engineers who can provide strategic direction, manage quality, and train internal resources on the tool.

The Impact of Artificial Intelligence on the Healthcare Industry

Artificial Intelligence (AI) is predicted to play a game-changing role in patient care. Let’s take a small example of its help in medical diagnosis. Imagine a scenario where a patient walks into a doctor’s office with symptoms indicative of several possible illnesses.  The doctor, to be sure, consults a digital assistant which scans a global database and comes up with a solution based on deep data analysis. The doctor goes on to prescribe further tests to confirm the prediction,  and here too, machine learning helps with comparing the images to the database and confirms the most likely cause of illness.  The doctor has just hastened patient care and with the help of accumulated intelligence has diagnosed the case. Not stopping there, the doctor introduces the patient to a chat-bot that explains the disease and its treatment. It schedules follow-up visits as well as any further investigations, if required. AI has just proved how invaluable it can be in patient care, by shortening the diagnosis to treatment curve.  Where time is of the essence, AI has proved how invaluable it can be.

Machine learning has brought AI to the forefront of healthcare and it is likely that its impact on diagnosing and treating diseases will be unsurpassed.  Recognizing this trend, a 2016 study by Frost & Sullivan, projects AI in healthcare to reach $6.6 billion by 2021, a 40 percent growth rate.  The study further confirms that AI will enhance patient care delivery by strengthening the medical imaging diagnosis process.    As an industry disrupter, AI will create real value for patients by supporting prevention, diagnosis, treatment, management and drug creation.

Technology experts predict that in the next couple of decades AI will be a standard component of healthcare – augmenting and amplifying human effort.  Its role will be as impactful and as quiet as the common X-ray machine.  It will also automate several health care tasks that are time-consuming and which require tons of unstructured data to be converted into intelligence.

While some of the innovations that we are talking about are futuristic in nature, AI has already quietly infiltrated this industry. It is already being used by healthcare players to manage billing, appointment fixing, and logistics planning.  To move into core clinical areas requires an amassing of data and that too has already begun.  With quantifiable data, diagnostics will become accurate and as a result indispensable in medical treatment.  Does this mean that we will see robot doctors in the place of human medical professionals?  Let’s leave that to science fiction movies for now.  What is more likely to happen is AI-enabled medical professionals.

To summarize, we can only imagine AI’s impact on saving human lives, going forward. For example, just imagine people in remote areas with limited access to diagnostics.  AI has just helped the local medical professional to remotely prescribe treatment, deliver medicines through an automated delivery system and prescribe telemedicine.  In a way, it has just helped to shrink the world.

Technology companies focusing on the healthcare segment are investing in Centers of Excellence where AI empowered healthcare IoT will bring about some dynamic changes, not to mention better control over existing processes such as supply chain, inventory management, equipment management, invoicing and drug development and reduce latency, lower cost and deliver operational efficiency. At Trigent, while we solve the problem of productivity, we remain focused on helping healthcare organizations take care of more people with less resources.  We do this by tapping our knowledge, experience and expertise in data and machine learning.