Connect with us

Science & Technology

Can State-of-the-Art Machine Learning Tools Give New Life to Household Survey Data?

MD Staff

Published

on

In 2014 the UN called for a data revolution to put the best available tools and methods to work in service of achieving the Sustainable Development Goals. Researchers at the World Bank have responded to that call by scouring the globe for the latest machine learning tools to transform our approach to tracking progress in the fight against poverty.

“Collecting household survey data on poverty is expensive and time-consuming, which means that policy makers are often making decisions based on old data,” said Asli Demirguc-Kunt, Director of Research at the World Bank. “Machine learning could drastically change the game, making poverty measurement cheaper and much closer to real-time.”

Machine learning is a field of computer science that allows computers to examine large bodies of data to identify patterns that data scientists would never find on their own.

Olivier Dupriez, a statistician at the World Bank, unveiled results from his and colleagues’ ongoing work on machine learning and poverty measurement at a recent Policy Research Talk. According to Dupriez, machine learning could make it possible to use a handful of easy-to-collect pieces of household information—such as whether a household has basic appliances or consumes a specific type of food—to predict the poverty status of a household. This would allow for poverty estimation based on simpler household survey instruments, and, hopefully, more effective policies to end poverty.

Data scientists have been developing the foundations of machine learning over the last half century, but its use has only become commonplace with the vast increases in computing power of the last decade. Today, machine learning is employed in everything from facial recognition technology to spam filters. In the case of poverty measurement, machine learning tools are employed to examine large national household survey datasets—which can be augmented with data from other sources, such as geospatial data—to develop models that can most effectively identify poor households.

From the outset, Dupriez and his colleagues faced a challenge: academia and the private sector have already created a multitude of machine learning tools, each with its own advantages and disadvantages. One tool may do a great job of predicting national poverty rates, but perform poorly when identifying specific households that fall below the poverty line.

Given the large number of options, the first stage of research has been exploratory. Dupriez and his colleagues have used large, comprehensive datasets on household poverty from Indonesia and Malawi to test four different approaches—with widely varying results.

The first of these applied 10 out-of-the-box machine learning algorithms to predict which households in Malawi and Indonesia are poor based on a handful of available variables. None of these algorithms proved to be a clear winner; different tools performed better on different metrics of poverty prediction. However, combining the results of these 10 out-of-the-box algorithms significantly improved predictive performance on most metrics.

The World Bank researchers also crowdsourced options through a global data science competition that attracted 6,000 submissions. The five winners from Portugal, Russia, China, and the Philippines faced the challenge of predicting household poverty status using anonymized datasets; they had no idea what country the data were from or whether specific households were classified as poor. Their models slightly outperformed the best of the 10 out-of-the-box algorithms used in the first approach.

Dupriez also employed an automated machine learning process, where the computer itself identifies the optimal way to build a model of poverty prediction. This approach requires a truly high-powered computer—in this case a computer with 32 processors worked for two days straight. The model produced via automated machine learning proved to be a disappointment, however, as it fell short on most quality metrics. But this may have been the result of the task assigned to it—the computer was directed to optimize only a single metric.

Instead, the most impressive results came from pairing experts and machines. Big tech companies like Google, Netflix, and Amazon have been innovating in the area of machine learning to produce content recommendation systems. Dupriez hired outside experts (Peter Bull and Casey Fitpatrick of DrivenData), who deployed an algorithm similar to the one Google uses for its Play Store. This algorithm produced excellent results on almost all metrics, making it a strong contender for additional investigation despite its complexity.

While the results so far are promising, the research is still in an early phase, with much more work needed to turn machine learning into a practical tool that can make poverty measurement faster and cheaper. Globally, data scientists need to commit to openness around data, software, and methods to make this a reality.

“The reason that our outside experts were able to build the best performing model is that they know a lot about what model works with what kind of data and what kind of challenge,” said Dupriez. “We need to develop a library of reproducible scripts and open data where researchers can find any kind of machine learning model, no matter who created it, so that we as a community can apply it to the most pressing challenges the world faces.”

World Bank

Continue Reading
Comments

Science & Technology

Geopolitics of 5G and the South Caucasus

Emil Avdaliani

Published

on

The age of 5G internet is coming to the South Caucasus. However, as the US-China geopolitical divide looms ever larger, the region enters the global competition where competing visions for 5G’s deployment could hamper the deployment of the new technologies. As a result, Georgia will edge closer to the US, Armenia – to Russia-China telecoms, while Azerbaijan will try to navigate between the two extremes.

A global competition is unfolding between China and the US. So far it has involved economic struggle and some aspects of naval rivalry. A new dimension needs to be addes – technological competition. This is reflected in the race which is underway to deploy next generation 5G mobile networks. In contrast to 3G and 4G (LTE – Long-Term Evolution), where China was largely relegated to the sidelines in the standards-setting process, now it has been heavily involved in the standards process for 5G—a sign of Beijing’s growing ambitions and global influence. 5G will be both quantitatively and qualitatively different from what has come before. For instance, as opposed to the hardware that drives traditional data networks, 5G networks’ primary functions will be software-based.

5G will be fundamentally different from its predecessor, 4G. Initially, vastly higher speeds will not be noticeable as it is likely that 5G will be used by network operators just to boost capacity on existing 4G networks and ensure a consistent service for customers. Ultimately, the speed capacity you get will be contingent upon the spectrum band the operator runs the 5G technology on and how much your carrier has invested in new masts and transmitters.

5G is also groundbreaking as it will usher in an extensive use of artificial intelligence (AI) to manage the network. Its ramifications are wide range: from military to civil spheres. And China has gained a significant advantage as it successfully moves toward commercial-scale deployment of its domestic 5G network in 2020. Consider the following example, in 2019 China’s Huawei signed an agreement with Russian telecoms on development on 5G networks in Russia.

According to some calculations, a full-scale introduction of 5G networks will take more than a decade. The process will be a complex one and the pace of its deployment across Eurasia and specifically in the South Caucasus will depend on a range of different factors such as carrier preferences, government regulatory policies, public and national security concerns, geopolitical alignments and costs of 5G infrastructure itself.

The development of 5G is taking place at the time as the world is becoming divided along economic and increasingly technological lines. The US builds its own bloc where Chinese technologies will have no major role. The same could be said about China. In this bifurcated world, small and techno-economically vulnerable states, which wish to gain access to the advanced technologies, will face tough choices which 5G network technologies to adopt. Small states across Eurasia are likely to face an increasing pressure from the US and its allies to abstain from using China’s 5G model, while Beijing will cushion its 5G offer in large financial benefits (investments, inclusion into the Belt and Road Initiative (BRI) etc.). Moreover, for small and relatively poor states Chinese technologies are more affordable than any Western competitors.

This brings us to the South Caucasus, which over the past several years has been gradually transformed into the frontline region between the US and China. Armenia, Azerbaijan and Georgia need 5G to boost their economic performance and relative technological backwardness. 

In Armenia’s case, with its booming IT sector, China’s low cost 5G provides a good opportunity to keep apace in the technologically developing world. Since Viva-MTS, a leading mobile operator in Armenia, belongs to the Russian MTS, which in turn partners with Huawei, Armenia is well placed to become one of the first states in the region to successfully deploy 5G technologies. Indeed, work has already begun, as the head of the department of technological development and resource management of telecommunications management of the Public Services Regulatory Commission of Armenia Armen Hunanyan said in one of his statements. 

Armenia is thus more likely to find itself in the middle of the US-China rivalry compared to Georgia or Azerbaijan. The latter two do not have similar Russian (by extension Chinese) influence in telecommunications. In fact, Baku made an interesting move in December 2019 when Ericsson and Azerbaijani communications service provider Azercell signed a three-year 5G Memorandum of Understanding (MoU) for the joint deployment of 5G projects. As a reminder, Ericsson is one of four companies in the world possessing enough technological prowess to supply the components necessary for building a 5G network. It is notable that US companies are absent from this hi-tech race.

For Georgia 5G will provide a much faster data download and upload speeds as well as much wider coverage from across the globe. With a chance of reaching 1Gbps, as opposed to presently provided 45Mbps, internet operations speed will be about 10-20 times faster. It could also help the country correct its overall technological underdevelopment in the provinces. The Georgian National Communications Commission announced that the introduction of 5G internet in Georgia will begin in 2020, but the pandemic and a general economic slowdown might change the deployment schedule. All in all, presently the telecom company Magti is arguably best poised to become a leader in the emerging industry.

Georgia also hopes that 5G technologies be applied to boost the country’s defence and security capabilities, especially in the light of the ongoing military troubles with Russia. Faster information flow will be crucial for developing more efficient defence capabilities along occupation lines near Abkhazia and the Tskhinvali region, where Tbilisi faces creeping occupation of lands and an endless string of kidnappings. However, the deployment of 5G might also bear risks as militarily developed states will be better positioned to use the 5G technologies to collect sensitive security and defence information from Georgia’s vital state sectors. On that front, deeper cooperation with US security services and the military sector will be crucial. However, there is a much larger problem Georgia, along with Armenia and Azerbaijan, could face due to the deployment of China-produced 5G. As the US-China divide grows and Washington is increasingly worried that Beijing might use this superiority for military and economic purposes incompatible with Western security practices, US pressure on South Caucasus states is likely to ramp up. Georgia will be under particular scrutiny as it positions itself as an ally of the US in the region and simultaneously seeks closer economic relations with China. In the long run Tbilisi may face the dilemma of having to choose between Washington’s support and the use of China’s technological advancements. 

With Armenia and Azerbaijan, it will be different to a varying degree. Security-wise Armenia is more related to Russia and its telecoms, Azerbaijan not so much. Moreover, as both are not openly aspiring to join Western political or military organizations, harsh US pressure on Yerevan and Baku might be less expected than in the case of Georgia. Still, a certain push from the US for a China-free 5G alternative will follow. It could hamper 5G deployment in the South Caucasus, the development which would once again indicate how global geopolitics is shaping the timing and level of the use of China’s or China-free 5G technologies.

Author’s note: first published in Caucasus Watch

Continue Reading

Science & Technology

Global transmission and control of Covid-19

Published

on

COVID-19 spreads mainly by droplets produced as a result of coughing or sneezing of a COVID-19 infected person. This can happen in two ways: One can get the infection by being in close contact with COVID-19 patients (within one Metre of the infected person), especially if they do not cover their face when coughing or sneezing. And secondly, the droplets survive on surfaces and clothes for many days. Therefore, touching any such infected surface or cloth and then touching one’s mouth, nose or eyes can transmit the disease. The most important factor in preventing the spread of the Virus locally is to empower the citizens with the right information and taking precautions as per the advisories being issued by Ministry of Health & Family Welfare. Preventive methods of COVID -19 include: Wash your hands often; Wear a face mask; Avoid contact with sick people; Always cover your mouth while sneezing and coughing. Several precautionary measures have to be taken in order to contain the virus because the risk of transmission will certainly increase with the lifting of the lockdown. We must recognise that this virus is going to stay on for some time and we have to make sure that at least for the next one year, we try and keep the virus as slowly moving as possible by physical distancing and other protective measures like masks and handwashing.

Transmission of the virus

We all know that COVID-19 spreads from others who have the virus. A person catches the virus through droplets expelled when someone infected with coronavirus coughs, sneezes or speaks. But, what happens in case of asymptomatic COVID-19 patients? Can they spread the virus? The World Health Organization, in a recently released video, answers similar questions related to COVID-19 transmission. In the context, Dr. Maria van Kerkhove, COVID-19 technical lead, says that the majority of transmission that is known till now is that people who have symptoms transmit the virus to other people through infectious droplets. “But, there are a subset of people of don’t develop symptoms. We still don’t have the answer to understand how many people don’t have symptoms,

Concept of Asymptomatic or Pre-symptomatic

Asymptomatic COVID-19 positive means a person has been tested positive but has no symptoms, and does not go on to develop symptoms, says Dr Kerkhove in the video. “A number of people are reported asymptomatic, actually may have mild disease. They may go on to develop symptoms. They may not quite register that they are sick. They can feel just a little bit unwell or under the weather, or fatigued. Some of those individuals we would classify as pre-symptomatic,”  Pre-symptomatic COVID-19 positive means that a person has not yet developed symptoms. Here we need to better understand what proportion is that contributing to transmission. This is one of the major unknowns. Explaining how asymptomatic people can spread the virus, Dr. Mike Ryan, EXD, WHO Health Emergencies Programme says, “For instance, if someone is in a nightclub, trying to talk to someone, and its too noisy, and you are too close to them, it’s like you are projecting your voice at someone. In this situation, if the virus is present your upper respiratory mucosa, then there’s every likelihood that you can project the virus.”Many people with COVID-19 experience only mild symptoms, especially in early stages of the disease. “It is possible to catch COVID-19 from someone who has just a mild cough and does not feel ill. Some reports have indicated that people with no symptoms can transmit the virus.

The current status

In terms of pandemic generation, we are still very much on the upward climb on this mountain, Dr. Ryan adds. “As some countries have shown, if you go at this with a comprehensive approach in a very systematic way, then there is enough stop ability for the virus.”.He also added  that as a society, we have some choices to make. If we can identify cases and their contacts, and we ask those contacts to quarantine themselves, we support them in that quarantine, then that can be a very successful way of both stopping the disease and avoiding large-scale lockdowns in future. Because, if one looks at the spread to people without history of travel or history of contact, certainly there are several such cases, he said. “But most of them are concentrated around the original points of entry of the foreign traveler’s or the travel routes of their contacts. So, these people who are describing it as stage 2 still are saying this is traceable local transmission, it is not unpredictable community transmission.

Therefore, we are avoiding the term community transmission. It is a matter of definitions and language; we need not debate that really. But  it should be recognised that community transmission has occurred in virtually every country which experienced this pandemic in a major form and India should also be prepared for it and act as though it is happening and take all precautionary containment measures. There is not only risk but actually threat of community transmission. According to him, nations in South East Asia, including Malaysia, and India in particular, have kept the COVID-19 death rates per million of the population low compared to countries where the pandemic broke out around the same time. The low death rate in India could be the benefit of multiple factors such as younger age group, more rural population, temperature and climatic conditions as well as the benefits the containment measures which preceded lockdown, and then got much more consolidated with the lockdown. Its quite possible that all of these factors have been helpful and we have seen that benefit. But we need to continue to consolidate that. There are some risk factors, when the lockdown opens there will be much greater mobility of people, there could be more widespread transmission of the virus, so we have to maintain as much as possible physical distancing, continue practices like wearing masks and hand-washing as precautionary measures.

However, these  are going to be difficult in overcrowded areas, especially slum areas. We will have to try and provide as much facilities as possible, particularly for elderly people and to people with co-morbidities, whether they can be provided temporary shelters elsewhere with good social cares. Fortunately, most of the infections are restricted to large cities and areas radiating around them. Referring to return of migrant workers,  care must be taken to see that they themselves will not be victims of the epidemic, and at the same they don’t infect others. But most important thing is to protect the rural areas (from COVID-19) because two-thirds of India is in rural areas, and the transmission of the virus is low there because mobility is low. Evolutionary biology of the virus says that when the movement is greatly restricted and its chances of transmission are greatly reduced, the virus actually can turn into a milder virus.

Continue Reading

Science & Technology

Entrepreneur hoping to make electric motorbikes a staple of Thailand’s streets

MD Staff

Published

on

Thai entrepreneur Soranun “Earth” Choochut, the founder ETRAN. Photo by ETRAN

Bangkok has an extensive public transit network, but its maze of small side streets—sois—means that a destination is often a hike from a transit stop. In a city of extreme temperatures and monsoon rains, even a short walk in the middle of the day can be very uncomfortable.

The solution for many living in Thailand’s capital are gas-powered mototaxis. This single-passenger mode of transportation can turn a sweaty 20-minute walk into a two-minute ride.

For Soranun “Earth” Choochut, founder and CEO of ETRAN, a sustainable mobility startup, navigating sois can be done just as well with electric vehicles. That’s why ETRAN has developed two models of electric bike, one designed for passengers and one a compact sport bike.

“We want to create a new age of public transportation that’s sustainable, clean and able to improve the quality of people’s lives in cities,” says Choochut.

The over 20 million motorbikes and scooters that whiz around Thailand’s streets and sois produce upwards of 18 million tonnes of greenhouse gases every year. Switching to electric would reduce that to “zero”, says Choochut.

ETRAN’s passenger bike not only provides cleaner transport, it is also helping address gender inequality. It is easy for a man, who is typically wearing pants or shorts, to hop on the back of a mototaxi. Women wearing skirts are forced to ride side-saddle, a more dangerous position. Without an internal combustion engine, ETRAN was able to slide the seat forward and leave a gap in the middle of its PROM model, allowing for a forward-facing sitting position for the passenger no matter the attire.

ETRAN also aims to reduce as much as possible its carbon footprint and plastic consumption.

Choochut’s efforts have received plaudits from the design world, winning the Red Dot design award twice. His environmental ambitions were awarded as well. He was one of the 2020 winners of the United Nations Environment Programme’s Asia Pacific Low Carbon Lifestyles Challenge.

With 3,000 units sold, ambitions to scale to 10,000 over the next two years, and partnerships with the likes of PTT Group and Bosch, Choochut wants to make sustainable transport a Thai economic success story.

“ETRAN can be the aspiration for others that proves that Thailand can invent something and scale like other countries and create passion and movement among entrepreneurs to create something bigger than themselves,” says Choochut.

UN Environment

Continue Reading

Publications

Latest

Newsdesk2 hours ago

Vietnam: New WB Support for Higher Education and Urban Development

The World Bank Board of Executive Directors today approved financing for two projects worth a total of US$422 million to...

Economy4 hours ago

Armenia’s inability to solve pandemic-related economic problems

According to data from the Armenian government, in 2019 the country’s economy grew by about 7.6%,which was the highest figure...

Terrorism6 hours ago

Attack on Pakistan Stock Exchange: A Fuel-to-fire in Southeast Asia

On June 29th, four armed men opened fire in the premises of Pakistan Stock Exchange, Karachi. All four terrorists were...

Central Asia8 hours ago

30 years of Nazarbayev’s foreign policy: What Kazakhstan can teach the world in the new era

The COVID-19 pandemic has completely shaken the very foundations of the world order that we were all accustomed to. The...

Reports10 hours ago

Major Impact from COVID-19 to Thailand’s Economy, Vulnerable Households, Firms

Thailand’s economy is expected to be impacted severely by the COVID-19 pandemic, shrinking by at least 5 percent in 2020...

Energy News12 hours ago

The Gambia: World Bank to Strengthen Access to Energy and Water

The World Bank Board of Executive Directors approved today a $43 million grant from the International Development Association (IDA)* for The Gambia’s Electricity Restoration and Modernisation Project (GERMP). The additional financing...

Middle East14 hours ago

Economic Crisis Does Little to Dampen Mohammed bin Salman’s Pricey Ambitions

When Saudi Finance Minister Mohammed Al-Jadaan announced austerity measures in May, including an $8 billion USD cut back on spending...

Trending