Implement and test an unsupervised machine learning algorithm: affinity propagation.
1.1 Reading
As first step, familiarise yourself with affinity propagation, for example by working through https://en.wikipedia.org/wiki/Affinity_propagation.
1.2 Task 1: Prepare a test dataset
Pick a small set of values that enable you to apply clustering using affinity propagation on paper.
Marking information: Up to 10 points: clarify why you believe your small dataset to be suitable for clustering.
1.3 Task 2: Apply affinity propagation on paper
Work through your small dataset from Task 1 to ensure you have known-good results.
Marking information: Up to 20 points: include information on how you sanity-checked your results as the number of iterations may considerably affect the precision of your final values.
1.4 Task 3: Implement affinity propagation in Python
You are now set to actually add the implementation. Note that you are expected to fully implement the mathematical operations instead of using a library function such as scikit or statsmodels.
Your implementation may make several assumptions about the inputs provided to it. Make those explicit in comments. Also, provide an estimate on how long execution of your algorithm will take dependent on the input values. Express this estimate as a function of the inputs, such as the number of clusters and/or the number of data points.
Marking information: Up to 50 points: 30 points for a correctly working Python imple- mentation, 10 points for describing limitations and assumptions of your implementation,
and 10 points for a description of the complexity of your algorithm.
1.5 Task 4: Create a test harness
Test your implementation via unit and integration tests. As part of this work, you may choose to compare to a reference implementation, such as sci-kit learn’s.
Marking information: Up to 20 points: identify suitable unit tests and integration tests.
Explain what coverage you expect your test suite to have.
The goal of this task is to implement and test an unsupervised machine learning algorithm, specifically Affinity Propagation. First, I will familiarize myself with the concept by reading through the Wikipedia page on Affinity Propagation (https://en.wikipedia.org/wiki/Affinity_propagation). This provides me with a good overview of how the algorithm works and its potential applications.
As first step in this process I plan create a small set of values that can be used as input for testing affinity propagation . To accomplish this ,I will use data from US Census Bureau’s 2017 population estimates which contains information about various characteristics such as age range , ethnicity etc… By isolating particular demographic groups within cities , I believe can then apply clustering using affinity propagation on paper to see if any patterns emerge . For example if there are multiple communities where majority individuals fit into a similar age range or belong same ethnic background ; then this could indicate that these neighborhoods have more commonalities than those around them.
Once I have my data set ready it is time put it through affinity propagation; however in order ensure accuracy results should be sanity checked via running program multiple times with different parameters each time until final values match up what was initially expected . Additionally, checking output at specific intervals during iterations may also help identify any issues earlier rather waiting until end calculate all necessary values.
appreciation and discourse creation. Because of this left hemispheric impedance, the issues in holding letter-sound affiliations may be made sense of by the singular perusing by means of the right half of the globe (Broca’s region which is overwhelmingly liable for appreciation) and not discourse creation.
It has been expected and adversely generalized that dyslexic individuals are not quite as scholarly as non-dyslexic people. This might be a result of troubles in school tests. For instance, inability to sort out scholarly realities successfully in tests might have a significant effect on their grade. Likewise, on the off chance that work is written in a disconnected design with oversights of words and expressions – basically on the grounds that the individual has forgotten about the thing s/he is attempting to communicate – can likewise have a negative effect on the result of the test. Notwithstanding, reality may eventually show that non-dyslexic students adjust an alternate technique for update that helps them to perform better in tests. For instance, update methods frequently applied by understudies like making records, practicing material, or verbally rehashing data into a recording device, would be a troublesome and distressing undertaking to a dyslexia student who experiences issues holding phonological data utilizing practice or reiteration or potentially issues with changing over phonological material into composed design. Moreover because of troubles with semantic data and memory, dyslexic people find connecting verbal marks to pictures ineffectual as a learning procedure.
Hence, it may not be that they don’t have a clue about the data expected of them in a test circumstance, it is more to do with their absence of skill in recovering and communicating their insight that is the deterrent. In this way it is significant in the instructive setting for dyslexia to be evaluated for since the beginning – so youngsters with dyslexia can learn various methodologies of learning data and be raised to their ordinary homeroom level.
Considering the proof above, apparently the etiology of dyslexia has its underlying foundations in the capacity of the human mind to get to its functioning memory. For instance, alluding back to a portion of the side effects showed in dyslexic people – for instance, troubles with records and reviewing data – it appears to be truly conceivable that parts of working memory or transient recovery of memory are not functioning as successfully as they ought to be. Natural proof likewise prompts the end that harm to the left half of the globe (the piece of the mind answerable for understanding) is normal in the dyslexic victim. Besides, the proof proposed above infers a neurological brokenness of the human cerebrum and the manner in which it gets to and uses the memory compartment. In this way could it at any point be expected that there might be a hereditary connection to dyslexia (see Byrne et al (2002) for help)? In any case, the help for the hereditary presumption of dyslexia is blended, and like a ton of twin examinations, jumbling results might be made sense of by natural improvement too, or more than, hereditary elements. Consequently, in addition to the fact that it is critical to perceive and analyze dyslexia since the beginning, the individual must be energized in the family and instructive setting to appreciate and create upon their frail language abilities. The mindfulness that parts of memory are a huge supporter of the abilities to learn of dyslexic endures will help families and tutoring to adjust to additional cordial approaches to instructing -, for example, multi-tactile showing strategies – so the youngster can get to other boost to set off their functioning memory.
Cerebrum imaging strategies uncover a scope of practical and primary cerebral irregularities of people with dyslexia. Cerebrum imaging studies have found the parts of working memory in the mind – the focal chief in the cerebrums, the visual-spatial sketchpad (‘internal eye’) in the right side of the equator and the phonological circle (‘internal ear’) in the left half of the globe (Carter, 1998). This might assist with making sense of why some dyslexics have unfortunate grapheme-phoneme transformation abilities – in that they experience issues changing over visual words into sounds. This might prompt the more unfortunate perusing of non-words, or words that are challenging to articulate. It has been refered to that the dyslexic students might utilize phonological codes in their functioning memory. Pickering (2000) proposes that dyslexic people have issues making an interpretation of visual data into phonological structure; this then influences their capacity to learn new words while perusing. This connects to investigation into Broca’s/Wernicke’s regions. In that harm to the Wernicke’s region which is liable for the capacity of recollections of the grouping of sounds that comprise into words – frequently creates unfortunate appreciation and discourse creation. Because of this left hemispheric hindrance, the issues in holding letter-sound affiliations may be made sense of by the singular perusing by means of the right half of the globe (Broca’s region which is transcendently answerable for perception) and not discourse creation.