Daily Tech Digest - November 27, 2022

Business Case – Why Enterprise Architecture Needs to Change – Part I

The solution to moving out of the “stone age” is to use a digital end-to-end approach for Architecture content (whether EA or SA), and provide openness and transparency across EA, project, and reusable component Architectures. Just like any digital approach to any business problem, the use of structured data is key. The best-structured data language for Architecture is arguably the ArchiMate notation which has a rich notation covering the depth and breadth of Architecture modelling, and also a rich set of connectors to link elements. ... Even if the new hire has significant experience in the given industry, the new organisation’s IT platform and processes will likely vary greatly from the person’s past experience. It takes several months or longer for new staff to accumulate enough knowledge about how the business and IT platform work to operate effectively without help from other staff and operate effectively. The cost of this knowledge gap is the new person delivering outcomes slower than other staff and consuming time of other staff unnecessarily by simply asking questions like ‘what systems do we have?’, ‘what does the business do?’, ‘how does system X work?’ and so on.


Why API security is a fast-growing threat to data-driven enterprises

API security focuses on securing this application layer and addressing what can happen if a malicious hacker interacts with the API directly. API security also involves implementing strategies and procedures to mitigate vulnerabilities and security threats. When sensitive data is transferred through API, a protected API can guarantee the message’s secrecy by making it available to apps, users and servers with appropriate permissions. It also ensures content integrity by verifying that the information was not altered after delivery. “Any organization looking forward to digital transformation must leverage APIs to decentralize applications and simultaneously provide integrated services. Therefore, API security should be one of the key focus areas,” said Muralidharan Palanisamy, chief solutions officer at AppViewX. Talking about how API security differs from general application security, Palanisamy said that application security is similar to securing the main door, which needs robust controls to prevent intruders. At the same time, API security is all about securing windows and the backyard.

 

Artificial Intelligence Can Enhance Banking Compliance

Technology has changed our society, and banks and other financial institutions have digitalized their operations at a rapid pace as well. However, the financial crime compliance units of these institutions still rely mainly on heavy manual processes. The banking compliance units’ key reason for their cautious approach in the utilisation of AI and automation has been uncertainty about technology. Do regulators approve machine-based decision-making, and is machine learning logic fair in identifying suspicious activities? However, there is a clear need for utilising technology in financial crime compliance. During the last number of years, Ireland has witnessed a rise in financial crime, with illegal proceeds making their way into the financial system, often from international sources. Last month, data from Banking and Payments Federation Ireland showed that over €12m was transferred illegally through so-called ‘money mule’ accounts in the first six months of the year. When compared to the same period last year, the quantity of bank accounts linked to the criminal practice in Ireland almost doubled to 3,000 between January and June 2022.


Big tech has not monopolized big A.I. models, but Nvidia dominates A.I. hardware

Interest in A.I. software startups targeting business use cases also remains formidable. While the total amount invested in such companies fell 33% last year as the venture capital market in general pulled back on funding in the face of fast-rising interest rates and recession fears, the total was still expected to reach $41.5 billion by the end of 2022, which is higher than 2020 levels, according to Benaich and Hogarth, who cited Dealroom for their data. And the combined enterprise value of public and private software companies using A.I. in their products now totals $2.3 trillion—which is also down about 26% from 2021—but remains higher than 2020 figures. But while the race to build A.I. software may remain wide open for new entrants, the picture is very different when it comes to the hardware on which these A.I. applications run. Here Nvidia’s graphics processing units completely dominate the field and A.I.-specific chip startups have struggled to make any inroads. The State of AI notes that Nvidia’s annual data center revenue alone—$13 billion—dwarfs the valuation of chip startups such as SambaNova ($5.1 billion), Graphcore ($2.8 billion) and Cerebras ($4 billion). 


Predictive Analytics in Healthcare

Clinicians, healthcare associations and health insurance companies use predictive analytics to articulate the probability of their cases developing certain medical conditions, similar as cardiac problems, diabetes, stroke or COPD. Health insurance companies were early adopters of this technology, and healthcare providers now apply it to identify which cases need interventions to avert conditions and enhance health outcomes. Clinicians also use predictive analytics to identify cases whose conditions are progressing into sepsis. As is the case with numerous operations of predictive analytics in healthcare, still, the capability to use this technology to read how a case’s condition might progress is limited to certain conditions and far from widely deployed. Healthcare associations also use predictive analytics to identify which hospital in patients are probable to exceed the average length of stay for their conditions by assaying case, clinical and departmental data. This insight allows clinicians to acclimate care protocols to observe the cases’ treatments and recoveries on track. That in turn helps cases avoid overstays, which not only drive up expenses and divert limited hospital resources, but also may endanger cases by keeping them in surroundings that could expose them to secondary infections.


How to Set Yourself Up For Success As a New Data Science Consultant With No Experience

The key is to know what you’re good at and focus on it. Going out on your own as a consultant is scary enough — ensure that you’re going to be marketing and using skills that you’re comfortable with. Having confidence that you can successfully produce results using your tools and skills of choice goes a long way to becoming a successful consultant. Additionally, do some market research to see where your niche could lay. While they say that data scientists should all be generalists in the beginning, I believe that consultants should focus on specializing themselves in niches that complement their skills and their alternative knowledge. For example, I would focus on becoming a data science consultant who specializes in helping companies solve their environmental problems — this would combine my specialized skills (data science) with my alternative knowledge and educational background in environmental science. Companies love working with consultants who have first-hand experience in their sector, so it can’t hurt to play to your strengths, past employment, education, or interest background.


The future of employment in IT sector

Whilst the companies keep up with the changing economic climate, what’s become undeniable is the war for recruiting good talent, now more than ever. There has been a significant change in employees’ needs and priorities. Cream talent is re-evaluating their careers based on aspects like flexibility, career growth and employee value proposition. Companies must therefore invest in ‘Active Sourcing’ to create a rich pipeline and not only recruit them but also train them for the upcoming 4th industrial revolution. It needs to invest in their skills and holistic development, not forgetting to create a safe, healthy work environment to retain the talent. As dynamic as it is, one cannot deny the menace of tech burnout. This blog describes it perfectly, ‘Tech burnout refers to the extreme exhaustion and stress that many employees in the technology sector experience. While burnout has always been an issue in many industries, 68% of tech workers feel more burned out than they did when they worked at an office.’ Technology is the most rapidly evolving industry with a challenging work environment.


On the Psychology of Architecture and the Architecture of Psychology

Most of our intelligence, however, consists of patterns that we execute efficiently, automatically and quickly. Some of these are natural elements, which are fixed: e.g. a propensity to communicate and use tools, to perform ‘mental travel’ — memory, scenarios, fantasy — and all of it based on pattern creation and reinforcement. Some of these elements may even be genetic (like basic strategies such as wait-and-see versus go-for-it you can observe in small children), but most of it is probably learned. All of this is part of Kahneman’s ‘System 1’. We learn by employing our capability to employ logic and ratio and our copying-and-being-reinforced capability — and while we do a lot more of the latter two than the former, culturally, we tend to believe that the reverse is true. Learning by reinforcement also includes learning by doing. Chess grand masters have very effective fast ‘patterns’ in the ‘malleable instinct’ part of their brains, and the difference between grand master and good amateurs is not their power of logic and ratio — calculating, thinking moves ahead — but their patterns that identify potential good moves before they start to calculate , and these patterns come from playing a lot of games. You also have to maintain your patterns: it is ‘use it or lose it’.


7 Common Data Quality Problems

Data inconsistencies: This problem occurs when multiple systems are storing information without using an agreed upon, standardized method of recording and storing information. Inconsistency is sometimes compounded by data redundancy. ... Fixing this problem requires the data be homogenized (or standardized) before or as it comes in from various sources, possibly through the use of an ETL data pipeline. Incomplete data: This is generally considered the most common issue impacting Data Quality. Key data columns will be missing information, often causing analytics problems downstream. A good method for solving this is to install a reconciliation framework control. This control would send out alerts (theoretically to the data steward) when data is missing. Orphaned data: This is a form of incomplete data. It occurs when some data is stored in one system, but not the other. If a customer’s name can be listed in table A, but their account is not listed in table B, this would be an “orphan customer.” And if an account is listed in table B, but is missing an associated customer, this would be an “orphan account.”


Building IT Infrastructure Resilience For A Digitally Transformed Enterprise

At a minimum, resiliency means having stable operations, consistent revenue, manageable security risks, efficient workflows, and an informed and agile employee base. Having visibility over the operating systems of network devices can reduce network downtime and open doors to further efficiencies. If a business is resilient, it can maintain stable network operations, drive down IT costs and deliver a more robust service at a lower cost. Overall, when businesses can dramatically lower IT expenses and have better visibility, they can expend resources on separate projects that improve the quality of service—a win for all. From a regulatory perspective, regulators now want to see everything documented. Take mobile banking, for example; regulators want to know everything, including what code is being used on which servers as well as which people and processes have access to which services. Intelligently automated network operations can allow enterprises to be better equipped to answer the questions that regulators ask, such as how they're validating and how often they're doing a failover. 



Quote for the day:

"A good general not only sees the way to victory; he also knows when victory is impossible." -- Polybius

No comments:

Post a Comment