Somali Insight Group - SIG

𝐇𝐨𝐰 𝐭𝐨 𝐞𝐟𝐟𝐞𝐜𝐭𝐢𝐯𝐞𝐥𝐲 𝐜𝐨𝐦𝐦𝐮𝐧𝐢𝐜𝐚𝐭𝐞 𝐰𝐢𝐭𝐡 𝐝𝐚𝐭𝐚 ?

Marki aad dooneyso inaad la wadaagto xog aad taxliilisey dadka ay quseyso (audience), waxaa lagama maar-maan ah inaad adeegsato muuqaalo (visualizations) ku haboon xogtaas kuna soo bandhigtaa qaab sax ah, hufan, dhameestiran misana la fahmi karo.

Waxyaabaha haboon inaad mar waliba hubisaa waxa ka mida:

🔑 Midabk - isku day mar waliba midab ku haboon xogta aad so gudbineyso inaad adeegsato.
🔑 Dooro visualization ku haboon xogta aad soo gudbineyso.
🔑 Sido kale hubi in aqristuhu ay u fududahay inuu durba fahmi karo ujeedka oo uusan jahwareer toona ka iman muuqaalka (visual) aad isticmaashay.
🔑 Hadi Pie chart isticmaasho, hubi in dhamaan qaybaha ka kooban yahay wadartoodu noqoto 100%.

𝐓𝐮𝐬𝐚𝐚𝐥𝐞 𝐚𝐡𝐚𝐚𝐧.
Waxaa dooneyso qaabka ugu haboon inaad xogtan soo socoto aad usoo bandhigto

42% shaqaalaha shirkada A waa dumar, 33% shirkadda B waa dumar iyo 30% shirkadda C waa dumar.

Adigoo ka duulaya qodabada horey ku xusan, muuqaal (visualization) noocee ah ayaad isticmaali laheyd ?

Pie chart 1 maxaa ka qaldan ?

30% + 42% + 33% isu geyntooda maaahan 100%.
Sido kale Company A vs Company B vs Company C waa population kala duwan iskuma dari kartid.

Si aan u saxno marka, shirkad walba xogteeda wa in hal pie chart lagu so gudbiyaa sida pie chart 2 ku dhigan oo kale.

maadaama qaladki aan saxnay laakin wali in wax laga badalo oo visualization professional ah oo tifatiran laga dhigaa.

Waxaa sameen karnaa in midabka aad ka badasho sido kale legend women vs men lagu badalo icons matalaya sida hoos ku dhigan oo kale.

Waxa hubaal ah pie chart 3 uu yahay mid ka haboon pie chart 1 &2.

Hadi aad rabto “wow” visualization ah waxaad adeegsata visual 1 iyo column chart, waa labada ugu haboon xogta aan horey kuso sheegnay hadi aad rabto si professional aad uso gudbiso .

Aniga shaqsi ahaan inta badan waxa isticmaala column chart ga oo waa qaabka dunida shirkadaha ugu waaweyn dhanka consulting ka sida BCG, PWC, Accenture ay uso bandhigaan xogta noocan oo kale ah iyo hayadaha caalamigaba.

Maqaalada so socdo ayan tusaaloyin kale oo aad u ifinaya isticmaalka statistical visualization ka kuso qaadan doona oo aad uga bogan doonta xogaha kala duwan iyo sida aad uso bandhigi karto.

-dhamaad

🔥 The differences between Supervised Machine Learning and Unsupervised Machine Learning Summarized in an image.

🔥 Do you know what Logistic Regression is? If you are unsure what it is or how it works, here is an infographic showing the detail!

🔥 Decision Tree Model is a base for many ensemble models, which many have used (e.g. Random Forest, Gradient Boosting, XGB). However, did you know how intuitively Decision Tree works? If not, you could check it out here!

Do you know how K-Means clustering work? or what are the purposes of this algorithm? If you are not sure, then this material could help you!

Increase in mean vs increase in variance vs increase both mean and variance explained in this graph and it’s interpretations.

🔥 Do you know how K-Nearest Neighbours or KNN work? or what are the purposes of this algorithm? If you are not sure, then this material could help you!

You want to visualize data then you will love ggplot2 in R. I use it daily in my routine analysis.

Create Elegant Data Visualisations Using the Grammar of Graphics A system for declaratively creating graphics, based on "The Grammar of Graphics". You provide the data, tell ggplot2 how to map variables to aesthetics, what graphical primitives to use, and it takes care of the details.

Need to brush up your stats concepts? This is my favorite FREE statistics course because it's thorough and intuitive with examples!
It covers hypothesis testing, bayesian methods, and so on.

Welcome! | STAT 415 Enroll today at Penn State World Campus to earn an accredited degree or certificate in Statistics.

Machine learning practitioners should learn A/B testing.

Machine learning projects doesn't stop once you build the model. How to you know your end users like the new model? Test it, Test it, Test it!!

To make a good decision, you need to understand randomization, sampling, and statistical testing to analyze the results.

A/B testing isn't just for product data scientist, it's for everyone in data science and engineering who build solutions for end users.