elijah101:
sir do you think that this avrasya city is a good investment ... it is going for 84k pounds

I don't know where the Avrasya city project is located. Irrespective;

For me, I often consider the following as my input parameters to predict if the investment is good or not when it comes to properties:

1. Location and significant attraction around the property: Think location first, and the property itself second; because the “right” property in the wrong location isn’t likely to be the right property at all. Proximity to amenities or facilities like University or Touristic beach/vacation zone is very ideal. A mid-to-long-term view regarding how the area is expected to evolve over the investment period must be an important factor to be considered.

2. Investment Purpose: Buy & Self-Use (Savings on rentals and value appreciation); Buy & Lease (Regular income and long-term value appreciation); Buy & Sell (Short-term: usually buy property under construction before selling at a profit on completion) or Buy & Sell (Long-term: value appreciation over a long period).

3. Return on Investment (ROI): The ROI must fall within 10-15%.

4. Fixes and Variable Expenses: Take note, an investment property abroad (Cyprus inclusive) is not a one-and-done purchase. There are expenses inherent in maintaining each property. Must times the developer don't tell you until you buy. An example, I pay 6500 pounds yearly for 8 apartments as "General upkeep costs (cleaning, landscaping, etc.)" which were not listed when selling the apartment. So you have to ensure you are ready for unexpected costs and your income from the rental must be able to fix it.

Best wishes

Shams08:
Yes you can even get up to 100% scholarship covering tuition fee only

..how posible is dat?

Admission Opens: Interested to Study?

For applicants interested to study at Eastern Mediterranean University (EMU): The admission process opens. EMU is the most ranked university in North Cyprus and with the highest international outlook and its degrees are internationally recognized with accreditations from notable bodies like ABET (USA) and ASIIN (Germany) and many more.

Undergraduate per year is averagely 4500$ (this includes 50% scholarship) and almost all courses are offered with top-level infrastructure and research facilities.

Private Universities in N.Cyprus: If you are interested in newer private universities like Cyprus Science/west Universities or Cyprus International University:

CIU: Cyprus International University is an English-language private university in Northern Cyprus and highly reputable among the private universities in North Cyprus. The Undergraduate fee before Corona Pandemic is 3100 euro/year but with the pandemic, the university board is currently offering 20% discount making the fees to be 2480 euro/year while the initial fee is now 1000 euro.


CSU: fees for undergraduate is 2600$/year and master program for the whole period is 2200$. The initial fees is 1500$


CWU: Here the fees for their available programs are attached here and the initial fee is 1400$.

If you interested in any of these universities, do not hesitate to contact me via my email or drop a message here.

Plz all d guru in the house. .keep us updated on when flight will resume in Turkey and n.cyprus. .....cus that's d most important thing right now. .....if not so, sch will have to offer us options of receiving lectures online. ..

haroldwilly780:
Plz all d guru in the house. .keep us updated on when flight will resume in Turkey and n.cyprus. .....cus that's d most important thing right now. .....if not so, sch will have to offer us options of receiving lectures online. ..

Yeah, there options to take the courses online if flights are not opened until Oct. That way, the university will offer a 30% discount (instead of 20%) until the students are able to Join the university physically.

Adminsion is on at CIU, initial sch fees is now 1000 euros, send your documents for adminsion at signature below.

Earlier this month the Turkish government announced flight operations to start on 28th May

With the pictures below, maybe it's becoming a reality, but on another thought, Turkey have been recording new cases everyday. Just in 24 hours they had 1,158 new covid19 cases

So guys let's keep our legs crossed, meanwhile if you need appointment, go book as they will resume tomorrow. And don't be in a haste to start your applications, unless you're ready for distant learning

Again if you have options to study in countries like US, Germany, Canada, Turkey, etc, please consider them. But if you know your whole aim is just to sturdy and acquire good academics, we will come you here

wip7:
Application of Artificial intelligence (AI) & Machine Learning to Laboratory/Field Data


I am a self-taught programmer with sufficient practical skills in high-level programming languages. Over the years, I have seen how AI is being used in various industries and I learnt this to inculcate in my laboratory research data and field results; this is awesome and has been very helpful.

In my lab, I have fed data from tourism sectors, waste management, pollution control, biomedical research and disease detection into ANN to discover intricate relationships or recognize complex patterns.

In particular, I have applied AI modelling to:

1. Predict Waste Generation Rates of Hospitality Industry (paper published in the Journal of Sustainability).

Here: For the first time, this study shows that municipal waste from hospitality facilities can be forecasted by considering measurable and effective parameters via an artificial neural (AI) network-inspired forecasting model. The field results showed that 4159.9 kg (recyclable: 58.5%, general waste:
23.6% and food residue: 17.9%) and 2063.4 kg (recyclable: 33.6%, general waste: 18.5% and food residue: 47.9%) of waste were generated during the peak and lean season from the 22 hospitality facilities investigated, respectively. While the ANN predicted that small and large hotels would produce 5.45 and 22.24 tons of waste by the year 2020, respectively.

2. To understand complex relationships in portable energy-storage devices fabricated in our lab (paper recently published in Journal of Electronic Materials)

Here: excessive strain on fossil fuel resources have ignited the global energy crisis; to curtail this menace, researchers have directed concerted efforts on energy storage devices. In this research, we used chicken bones to build portable energy storage supercapacitor and applied ANN to understand the complexities, thus shortened our experimental work by 69% and we are able to understand what parameter improves the energy storage ability of the supercapacitor.


Artificial neural networks (ANN) are one of the main tools used in machine learning. ANN is brain-inspired systems inter-connected layers which are intended to replicate the way humans learn. ANN is an excellent tool for discovering patterns which are far too complex or numerous for a human programmer to extract and teach the machine to recognize.

Below are some of my works where ANN have been used.

Diagnosis of breast cancer by AI: Here, AI was applied to assess whether a lump in a breast could be malignant (cancerous) or benign (non-cancerous) from digitized images aspiration biopsy.

Interested to study in TRNC? don't hesitate to contact me.

To be continued!

Why are you still using sigmoid ø as your activation function for your hidden layers? Isn't this outdated?

.

Near east Don reduce dia own too abi ??

etrange:
Why are you still using sigmoid ø as your activation function for your hidden layers? Isn't this outdated?

Sigmoid is never outdated, it depends largely on your data type and nature of machine learning/ANN application you want to use your data for

Machine learning or Artificial neural network applications are majorly two: approximation and classification applications.

There are several activation functions for the hidden layers but here the most common activation functions:

Linear activation function : This will not really make sense for classification application (in hidden layer) since the output of a perceptron with linear activation function is simply the combination of that neuron. It is like a=b. The linear activation function is very used in the output layer of approximation neural networks.

Hyperbolic tangent activation function: The hyperbolic tangent is widely used because it has a monotonous crescent function which exhibits a good balance between a linear and non-linear behaviour. Interestingly, it has a sigmoid shape and varies in the range (−1,1) making it very useful in the hidden layers of neural networks for approximation and classification tasks.

Logistic (Sigmoid) activation function: A logistic function is a common S-shaped curve (sigmoid shape) . If you look at the shape of the Logistic (Sigmoid) function, you will see that it extends within (0,1). This is a good property because the outputs here can be interpreted in terms of probabilities. Therefore, it is widely and frequently used for binary classification tasks.

Rectified linear (relu) activation function: This is a non-linear activation function and is more computationally efficient than other non-linear activation functions due to its simplicity, and very used in the hidden layers of neural networks for approximation and classification tasks.


ANN Classification tasks that use Logistic (sigmoid) activation:

---Breast cancer diagnosis: For this class of applications, the neural network is composed of scaling layer, perceptron layers and probabilistic layer. Take note the perceptron layers will use logistic hidden layer and a logistic output layer since the logistic function ranges from 0 to 1, the outputs from that layer can be interpreted as probabilities (either the person has breast cancer (1) or the person does not have breast cancer (0)).

----Detect forged banknotes: This is also a classification application where the ANN can predict via a binary classification as either (forged banknote (1) or legal (0)). So the task will be to model the probability that a banknote is fraudulent or not, as a function of input data features.

----Detect Fraudulent (spam) email from Yahooyahoo guys: Google and related companies use machine learning model for this application and its majorly a classification tasks (meaning it will use Logistic activation (sigmoid) or ReLu or Swish etc). For Google, it is now highly advanced to the point that it can detect and filter out spam and phishing emails with nearly 99.9 per cent accuracy. It either classifies the email as spam (1) or real (0)...this is sigmoid activation.

.....Many more applications.

Meanwhile, each activation has its advantages and disadvantages. For instance, ReLU is computationally efficient (allowing faster convergence of the network) and is non-linear; its major disadvantage is that when the inputs approach zero the network finds it difficult to perform backpropagation and cannot learn anymore. Google uses ReLU for many of their AI applications but some of the recent works from then seems to favour Swish which is a self-gated activation function and they say it performs nearly 1% higher in classification than ReLU. Sigmoid/Logistic gives clearer predictions and prevents Jumps in output values but it is computationally expensive especially if you have a very large dataset and also has varnishing gradient problem (making the network to be slow to learn).


You can always play with many activation functions, the entire goal is to reduce your errors, advance your classification or prediction level very high so as to build a model that can at least give nearly 99.9 accuracies. For me, I use various and I play with various activation functions just like I play with a various training algorithm.

I hope this answers your question.

wip7:


Sigmoid is never outdated, it depends largely on your data type and nature of machine learning/ANN application you want to use your data for

Hhmm...

Why do you have up to 3 hidden layers with all those nodes? Isn't that prone to over-fitting?

haroldwilly780:
Near east Don reduce dia own too abi ??

Yes, for undergraduate studies

Bravo!!! Turkish embassy is the first to start Visa application service and appointments today.
They took the bold step, other embassies will follow suit.

etrange:


Hhmm...

Why do you have up to 3 hidden layers with all those nodes? Isn't that prone to over-fitting?

Very interesting question.

Although the presence of more hidden layer makes training of the network a bit more complicated; however, no rule to as how many hidden layers a model should have. Practically, the more the hidden layers between the input and output layers turns the Perceptron into a universal approximator, which indicates that your network may be robust and capable of capturing and reproducing extremely complex input-output relationships.

If you have a very complex relationship, such a task may be accomplished faster with higher accuracy with Deep Learning Neural Network and deep learning can have as much as even 7 hidden layers.

A specific case, I had a project to design a model that can be used for a greenhouse farm project. The goal is to optimize the farm yield and notify the owner of changes in conditions (physical parameters) that may lead to issues. In this project, independent variables were 20, instances were 20,500 (quite large). Here, 11 training algorithms were used to train the dataset, the results of the training algorithm can be seen attached. LMA was the best training algorithm and it suggested 11 nodes of 3 hidden layers. So I did not choose the number of hidden layers but the training algorithm selected that considering the fact that am trying to minimize the errors.

etrange:
Isn't that prone to over-fitting?

Yeah, over-fitting is a common occurrence. This was eliminated by normalizing my data and I also applied two regularization methods (LI/L2 regularization). But from my experience, overfitting often occurs not because of the number of hidden layers but when the model tries to predict a trend in data that is too noisy.... and this can be eliminated by data augmentation.

Admission is on going at EMU, Near East university and University of Kyrenia, Apply Now, Check my signature for Admission!!

Emu Don reduce dia own tuition ??

haroldwilly780:
Emu Don reduce dia own tuition ??

Maybe they will maybe they wont

Happy Eid to All

As you all celebrate, may the joyous occasion of Eid spread joy, peace and happiness to your heart and across your home.

Prof. WIP

Prof wip, I sent a message to your email recently, am yet to read from you.

Kj40:
Prof wip, I sent a message to your email recently, am yet to read from you.

Sorry just replied all my email ...

Prof Wip, I graduated last year (Bsc) and didn't apply for the April Msc session because they haven't released our Bsc certificate. I don't want to wait until next year as school is still on lockdown because of the virus and I am not sure they will be issuing Bsc cert anytime soon, can i apply for Msc with my result transcript (100-400lvl) ? Your thread has been of immense help to me, thank you very much.

biggeek007:
Prof Wip, I graduated last year (Bsc) and didn't apply for the April Msc session because they haven't released our Bsc certificate. I don't want to wait until next year as school is still on lockdown because of the virus and I am not sure they will be issuing Bsc cert anytime soon, can i apply for Msc with my result transcript (100-400lvl) ? Your thread has been of immense help to me, thank you very much.

Do you have statement of result that you graduated?

If yes, then I can apply for with your transcript and the statement of result. If not, if you can get a letter from your university stating you have graduated then that will suffice pending when you are able to supply the original certificate.

I need an important information on cyprus international university. For example a course is 4 years does that mean i would have to spend the entire 4 years like Nigerian universities?

Can i fast track my schooling by increasing my load through summer programs thereby reducing the number of years to 3 years or even 2.5 years I heard its possible but i need confirmation.

zinaunreal:
I need an important information on cyprus international university. For example a course is 4 years does that mean i would have to spend the entire 4 years like Nigerian universities?

Can i fast track my schooling by increasing my load through summer programs thereby reducing the number of years to 3 years or even 2.5 years I heard its possible but i need confirmation.

Yea it’s possible to take courses available in the summer to reduce the number of years to finish 4 years course to like 3 in total and you might just go with the normal fall and spring semester for 4 academic year

Hello, I've just been admitted to study business administration in Cyprus west university, my question is, is it possible to switch schools once I'm there? And if not, is it possible to secure an admission into emu or near east university with a waec result with four credits and a nabteb with one credit?

jidoex66:

Yea it’s possible to take courses available in the summer to reduce the number of years to finish 4 years course to like 3 in total and you might just go with the normal fall and spring semester for 4 academic year

Thanks alot for the info. But does this attract any extra cost excluding the school fees?

wip7:


Do you have statement of result that you graduated?

If yes, then I can apply for with your transcript and the statement of result. If not, if you can get a letter from your university stating you have graduated then that will suffice pending when you are able to supply the original certificate.

Thank you for your help sir, I just contacted somebody who will be able to fetch me my 100lvl-400lvl transcript but I can't get statement of result or letter until school resumes, I can provide access to my student portal as well. Naija just has a way of delaying somebody's life and I don't want to wait anymore