₦airaland Forum

Hushpuppy family. Too many of them like MTN, everywhere you go.

emmanuelewumi:
Ambode's project

Haha, una no de tire. Abode sef remain inside with all de humiliation. Na him make Tinubu a progressive

Ipob won't get over theirs, whereas Ijaw has long ago got over it.

They are naturally dirty policemen. I have one in my former church, a building of 6 flats, 3 BR each, was confiscated from a robber, do you know these people turned this building to something else?

wip7:
first, corrosion of pipelines is caused by many reactions (redox) in a corrosive environment like when the pipeline reacts with another substance such as oxygen, hydrogen, SO2, an electrical current or even dirt and bacteria.

To collect data for corrosion, you have to first know the type of corrosion and that will help you to determine what causes that type of particular corrosion and then you can collect data for it. Type of corrosion are Uniform corrosion, Galvanic corrosion, Pitting corrosion, Stress corrosion cracking, Crevice Corrosion, and Microbiologically influenced corrosion (MIC).



To build a Machine learning model to predict corrosion occurrences; first, identify the type of corrosion you want to predict. Some of the data that will be needed to be collected will include: structural data, environmental data, and metallurgical data which can be some of these pH of the environment, temperature, pipeline composition (metallurgical factors), welding conditions, air/water pollution data.

For Microbial corrosion or MIC: this can be caused by bacteria, fungi, and microalgae. The aerobic bacteria like Thiobacillus thioparus, Acidithiobacillus thiooxidans and Thiobacillus concretions are the three major bacteria causing corrosion. So you have to identify the concentration of these bacteria in that environment, nature of reactions they cause, do they make the pipelines to smell like toxic hydrogen sulfide or the pipeline coating gone and looks more like lightweight graphite.....all these info will help you to identify exact bacteria and then come up with parameters to collect data from.

Meanwhile, there are valuable published papers on machine learning with corrosion occurrences.

God bless you, sir. You made my day. MIC will be suitable for this. I was thinking to include temperature, concentration, the pressure to this initiative but your explanation has simplified it to me now. This area hasn't been explored. Though, several types of research have been conducted in different methods like acoustic, radiation wavelength, etc for prediction. Thank you once again. consequently, I may ask more questions.

wip7:
Are you talking about corrosion inhibition or predicting corrosion occurrence?

To build a valuable AI model; first, you need to know the aim of the model and who will use the model. If your model will predict the occurrence of corrosion for a particular company or particular product. Then you have to collect data on factors that often cause corrosion from the company if you are building for the company.

For example, I am building ML model for Air Pollution Monitoring that will be a web-based in my city and can be useful in the country; what I did was to design my own device (via Arduino board) to collect concentration of the target air pollutants like Carbon monoxide, O3, PM10, Lead, Nitrogen dioxide, Ozone, Particles, Sulfur dioxide, etc. So we have 15 of the devices in different locations within the city and it collects data daily and for 3 months we started building the model and its ongoing project. My colleague have published part of this work recently in the Journal of Atmospheric Pollution Research.

However, my intention is to build a web page that gives air pollution index and forecast just like weather forecast sites.


The data can be from 5000 upwards depending on how much you want, the more the data the more the model can easily detect patterns.

Advantage of AI/ML and large dataset

1. Fast to spot a trend: Machine Learning can review large volumes of data and discover specific trends and patterns that would not be apparent to humans.

2. ML Self and continuous improvement : As the ML model gain experience, it keeps improving in accuracy and efficiency without human intervention so as to make better decisions. For example, if you want to make a weather forecast model. As the amount of data you have keeps growing, your ML model learns to make more accurate predictions faster.

Thank you, sir. Yeah, predicting the occurrence of corrosion in underwater pipelines caused by marine growth. I am not building for any company but to teach myself though. Who knows it might end in patent hahaha. I appreciate your response so much.

Is it the organism population that will become dataset.? Sorry, for too many questions. I am still a novice to ML.

WilliamsDeGreat:
Inbox me your CPC I like to know who you be. Since you know more about BAN

You be a real empty skull. He should inbox you his Amotekun �

wip7:
Each student comes with a different set of skills, experience, exposure, knowledge, and drive. For me, I personally don't accept every student that wants me to supervise them. I tend to first ensure the student is actually open-minded, really ready, and have a set of background skills.

So regarding your question: I have a wide research area and when students come I want to listen to what they want to do. So I evaluate their ideas and proposal according to these steps:

1. is the research worthy? is it a Ph.D. or master level research?

2. Level of knowledge and practical skills my student have?

3. What are the innovations and significant contribution the proposed research idea will generate and the impact?

4. I prefer a research idea that is not bound just to my lab but must be reproducible internationally and not necessarily 100% under my lab parameters even though if it's directed largely to solve national issues.

5. I prefer a research idea that has an element of lab-to-market.

6. I check, do we have the equipment, material, and required facilities to conclude the research within the allowed time-frame.

7. If we currently, don't have the equipment, do I have a network that can help with the facility or can I switch the student to another country for some time or can I purchase the needed facility to support the research? So I spend time on this part.

8. If this research is concluded, can we publish the data? who will benefit and what significant contribution are we going to attain.

So if my student's idea falls positively into the above-listed, then we start the research work based on the student ideas under my guidance.




Meanwhile, when this student came, she has no idea how to do this practically in fact she has less English and less lab experience but she has a worthy idea. Now she is doing perfectly great. She now teaches me how to grow potatoes and vegetables without soil just with a UV lamp. I have used the data into AI (artificial intelligence) model and we have fabricated a device to collect data straight to our cloud and that's really making a valuable contribution. Later I will post some of the publications on slow-release fertilizer hydrogel we did.


I hope this answers your questions.

It has answered even more than expected. During my master's degree, I have a lecturer that loved dictating to students what he want to extend that many of these students couldn't finish their thesis on time. It was horrible to some of these students to extend that some have to write for extension of submission because meeting the time of submission wasn't visible.

I will also like to ask you some questions about artificial intelligence relating to corrosion. How do you think the dataset in corrosion should look like? Is corrosion results from various companies or collection of individual results a dataset? You have inspired my desire to use machine learning in corrosion.
Thank you, sir.

wip7:
I think it depends on each university's rules and regulations. Each university often has a graduate institute office that coordinates and provides details regarding the thesis writing and formatting. In modern and standard schools, such information is often placed on the university website.

In general, the thesis does not have maximum nor minimum provided the key elements are all included proportionally.

I remember one of my classmates (or senior) we were nearly graduating at the same time during our PhDs;

---her thesis was like 350 pages with just one scientific publication as a requirement to graduate Ph.D. (her supervisor loves heavily paged thesis, she believes that means worthy and valuable and with such heavily paged thesis she said the student is highly competitive).

----On the other hand, my Ph.D. thesis was 40 pages and graduated with 25 scientific publications (although when my Ph.D. supervisors saw my thesis, they tried to convince me to increase the page so the Jury members won't say its too light.........for me I don't see any sense in that because I only summarized all my publications........I convinced the major supervisor that spending 3 years in lab/field with 25 publications and another with 7.5 years for Ph.D. with 1 publication even blind person will know......so no further comments). During the submission at the graduate institute, the director said the thesis appears light and fewer pages, I told him my supervisors and Jury members did not complain either myself... so......!


For the Ph.D. students I have supervised; I believe in quality, passing experience, training, and exposure to my students rather than giving me 1000 pages thesis and after graduation, they are not worthy or their CVs cant competes favorably globally. The university needs just 1 reputable publication to graduate, but my rule for my Ph.D. students is a minimum of 10 paper publications, some meet and exceed while some meet just 5 ......

Anyway, So check with your university and supervisor advice.

Sir, I want to know if you allow your research student to use their own idea or you asked them to do everything based on your own, not theirs.

Tinubu can now take a bow.

Why can't we have an Engineering school like law school where you write professional exams after first degree than all this money ventures by the board with no meaningful impact.

Mokason28:
Seconded... from the post above No body cares... IZE IYAMU is a thief according to Oshomole

If IzeIyamu is a thief, obaseki is a robber.

hustla:
I also have interests in knowing about the church of Satan, the 7 books of Moses, Christianity and Islam other than what we were told, Astral travel amongst many many other things most people don't like

I watch Forensic files/ First 48, I love documentaries which show me more insight into the life and how things came.to be. While others want to watch BBNAIJA, I prefer CI and History channel

If them no train you to seek knowledge and ask questions, na your problem be that, not mine

If you truly have an interest in all this rubbish, there are books you can also read, not coming to disturb for hearsay or story by moonlight.

fatymore:
How much is 200k to him...
He can afford that

You sure?

Ishilove:
You want to go by road for international flight? With Agofure or GIG?

Many of them don't read apart headlines.

nairavsdollars:
very preposterous. arriving three hours for a 30 minutes local flight

Relax your nerves, it is for an international flight.

AlabiJ:
Edo no be Rivers. Edo no be Abia. Edo na Edo.

haha Edo na Italo

dheilaw1:
Does he have the power to order the removal of mace?

And it is still a demonstration of crazy people "Democrazy persin"

Ayopercent:
Good afternoon house, please can someone recommend a good and free plagiarism checker that I can subscribe to online...

check www.turnitin.com and www.scribbr.com

Turnitin is the best pool of plagiarism checker but I don't know if you can subscribe individually.

xcolanto:
Good day all, it's really nice to locate a thread like this for the potential and ongoing Ph.D. research program. I currently want to enroll for a PhD and I was told I need a 3.50 GPA. I graduated with a master's in the UK with merit, my transcript says I got 180 points. How can I convert my merit to GPA to know if I qualify for the Ph.D. program? Thanks.

Since your average falls within the merit % you should qualify above 3.5 GPA. Merit is between 60%- 69%

ogbeking:
Change your vocabulary Idris! Goner since last year. He's still goning.

Change your spelling, it is ginning.

HIGHESTPOPORI:
Even Buhari, IG of Police, some Apc governors and all PDP governors are behind Obaseki. Police taking over the complex is to prevent thugs imported by politicians to attack the complex.

Machinery is insiders not all these outside spectators.

Truvel:
U Don't Av 2 Assume Like Dat Since U Don't Av Any Proof Of Ur Comments.

First impression matter bro.

Seanixking:
He is not lying. Na just some person dey carry am for head.

If you like carry am for heart and hand, rival wey carry am for head go offload am, both the heart and hand.

Truvel:
Don't Judge A Book By Its Cover. A Person May Be At Dat Level & May Possibly Not Kill Anyone.

When you cannot judge a book cover than its content, it matters that you present yourself both in appearance and character.

LaboPolitics:
No be every meat hunter dey chop.

Instead, the meat will hunt the hunter. Even, If it was easy all animals won't enter the zoo.

When you hear someone shouting "rugged you" if you watch very closely you will find out that he is dumbass, ode exponential.

Ambiitiouscent:
You dey mind those ones. Drop number, drop location, drop picture.

Everyone believes himself to be the Rambo in the story until reality sets in.....smh

How immature can someone be?

Both think they can rumble with the faceless threat. Babies online, indeed. Exhausted two pages for nonsense rants.

LaboPolitics:
I once read a piece of news a few years ago, where some kite guys were killed in one of the tertiary institutions in the SW, I started wondering how the kites got themselves into an occultic mess? I thought they only drink, have fun and gyrate.

I hear they even have names they call violent cultists 'Nkutumbali' or 'Ayamalele'. I'm still confused.

Don't be confused, its formation is relative to all cult formations. The only difference is that not too violent and the orientation still remains African culture, though it has become hideouts to various groups.

Deffjam:
jeune homme soit très prudent

vous êtes prudent ami. Il se comporte grossièrement

invectives:
you carry mode9 cry reach here??

really mode9 to cool nerves

urchcoded:
This thing no heard nau. You just need a computer that has a good camera and data. Abi e get any other thing?

It is not "hard" but difficult for public universities.