You are reading the article Unique Data Visualization Techniques To Make Your Plots Stand Out updated in December 2023 on the website Daihoichemgio.com. We hope that the information we have shared is helpful to you. If you find the content interesting and meaningful, please share it with your friends and continue to follow and support us for the latest updates. Suggested January 2024 Unique Data Visualization Techniques To Make Your Plots Stand Out
Andrews Curves for Data VisualizationAndrews Curve is a useful plot for visualizing patterns in multidimensional data. The concept Andrews Curves was developed by statistician David F. Andrews in 1972. Andrews Curves are created by defining a finite Fourier Series which is an equation of sine curves. This allows us to visualize the difference in data. The equation is given by:
where x refers to each of the dimensions in the data and the value of t ranges from -π to π. Learn more about Andrews Curves in Python Here and Wikipedia.
Let’s create the Andrews Curves in Python using Pandas.
Importing the Libraries
import pandas as pdImporting the Data
df = pd.read_csv("iris.csv")The dataset has been downloaded from Kaggle.
Plotting the Andrews Curves
plt.figure(figsize=(12,6)) pd.plotting.andrews_curves(df, 'Species', colormap='YlOrRd')Here, we passed our dataframe df into the .andrews_curves() method of pandas.plotting. We also passed our categorical variable and the colormap for the Andrews Curves.
Putting it All Together
Python Code:
On executing this, we get:
Source – Personal Computer
Raincloud Plot for Data VisualizationRaincloud Plot is a unique Data Visualization technique introduced in 2023. It is a robust visualization technique that combines a violin plot, a box plot and a scatter plot. Thus, one can see a detailed view of raw data in the single plot. This plotting style makes the Raincloud Plot better than any of the charts it is made of alone. Learn more about Raincloud Plots here.
4) “Thunder”, a pointplot connecting the mean of the different categories (if pointplot is True)
Installing the Libraries
pip install ptitprince
Importing the Libraries
import pandas as pd import matplotlib.pyplot as plt import ptitprince
Importing the Data
df = pd.read_csv("iris.csv")
Plotting the Raincloud Plot
plt.figure(figsize = (12,8)) ptitprince.RainCloud(data = df, x = 'Species', y = 'Sepal.Length', orient = 'h')import pandas as pd import matplotlib.pyplot as plt import ptitprince df = pd.read_csv("iris.csv") plt.figure(figsize = (12,8)) ptitprince.RainCloud(data = df, x = 'Species', y = 'Sepal.Length', orient = 'h') plt.show()
Source – Personal Computer
Calendar Heatmap for Data VisualizationCalendar Heatmap is a unique visualization technique to visualize the time series data. It creates a horizontal pallet of squares, each resembling a day of a year. One must have seen a similar palette in their GitHub profile for all the commits for a year. Since it’s a heatmap, each square has colors of different densities based on the values or weights for that day.
In Python, we can create a Calendar Heatmap using a library called calmap. Let’s build a Calendar Heatmap.
Installing the Libraries
pip install calmap
Importing the Libraries
import matplotlib as mpl import calmapImporting the Data
df = pd.read_csv("currency.csv")
Getting Data Ready for Plot
df['Time'] = pd.to_datetime(df['Time']) df.set_index('Time', inplace = True)Here, we converted our Time column to DateTime type. Next, we set the Time column as Index using the .set_index() method which will help plot the Calendar Plot.
Plotting Calendar Plot
calmap.calendarplot(df['2023']['GEMS_GEMS_SPENT'], cmap = 'OrRd', fig_kws={'figsize': (16,12)}, yearlabel_kws={'color':'black'})Here, we used the .calendarplot() method of calmap. In the calenderplot(), we specified the arguments as dataframe for the year 2023 for column ‘GEMS_GEMS_SPENT‘, cmap as ‘OrRd’, figure keyword arguments fig_kws for specifying the figure size, and yearlabel keyword argument yearlabel_kws to specify the colour of the year label at the left side.
Putting it All Together
import matplotlib as mpl import calmap df = pd.read_csv("currency.csv") df['Time'] = pd.to_datetime(df['Time']) df.set_index('Time', inplace = True) calmap.calendarplot(df['2023']['GEMS_GEMS_SPENT'], cmap = 'OrRd', fig_kws={'figsize': (16,12)}, yearlabel_kws={'color':'black'}) plt.show()On executing this, we get:
Source – Personal Computer
ConclusionsIn this article, we learned and implemented three unique visualization techniques to take our data visualization game to a next level. Unique visualization techniques are eye-catchy and gain attention from the viewers. Learning new techniques also helps in developing our skillset. One can try playing with arguments of the methods discussed above to build more robust and beautiful plots. With time, a lot of new visualization techniques are being developed, one should keep trying to learn them to create more accurate plots based on the data.
About the AuthorConnect with me on LinkedIn.
For any suggestions or article requests, you can email me here.
Check out my other Articles Here and on Medium
You can provide your valuable feedback to me on LinkedIn.
Thanks for giving your time!
Image SourcesThe media shown in this article are not owned by Analytics Vidhya and are used at the Author’s discretion.
You're reading Unique Data Visualization Techniques To Make Your Plots Stand Out
Hiring In 2023 Is Hard: Here’S How To Make Your Job Posting Stand Out
Were you, like so many of us, thinking that with the pandemic under better control things would start getting back to normal—where higher unemployment drives more applications? One would think this would be the case…but the data tells us otherwise.
If you follow the news, you know that the unemployment rate is still high, but for some reason, the “normal” logic doesn’t seem to apply anymore when you post your jobs. In fact, applications have been lagging and dropped 14% in the second half of March compared to the first half. What happened to high unemployment = more applicants. What is going on?
In this post, we’ll explore:
The data on recent hiring trends
Job posting basics you need to master
How to make your job posting stand out
Hiring trends and how they’ll impact your 2023 job postingsIn 2023, we saw record unemployment due to the pandemic, which led to fewer job openings across the US. We’re seeing that there are now a lot more job openings, but many companies are seeing fewer applicants than they did pre-pandemic.
Here’s what we know:
There are more job openings now than a year ago, even just a month ago, and we still can’t get enough applications in the door…
You are not only competing with your usual competitors in your area and/or industry, you are now finding you are competing against jobs across the country offering remote work, sign-on bonuses, higher than average hourly wages, and other enticing benefits.
Image Source
Here’s what this tells us for job postings and hiring in 2023:
There will be fewer applications for every job opening
The cost per hire is going up
Hourly rates is/will be going up*
*The hourly wage jobs are particularly hit by the jobs trifecta consisting of, perhaps not in equal parts, lingering COVID fears, the lack of trusted and safe in-person child and/or eldercare, and government regulations.
Related: Find out how to overcome top recruiting challenges here.
Mastering job posting basicsSo, what to do about these trends? Here are a few of the basics you need to master for a job posting that stands out.
Are your job postings fresh?When did you last update your job postings? Check your career page, ATS, and online listings, and make sure your job postings are less than 30 days old.
Are you expecting hires within days?Did you know the average time to hire is 42-45 days or more depending on the needed skill set? Make sure you plan to run your recruitment marketing campaigns for more than 30 days. With an average time to hire of 42-45 days, your best bet is 60-90 days ad runs, and that timeline may need to be longer. For example, if you are looking for a CDL driver, we see 80+ days to hire.
Image Source
Are your job listings easily found–everywhere?Are they ranking on Google, found on all the social media channels, and are your jobs targeting and retargeting qualified candidates? If not, it may be a sign to refresh your current recruitment marketing strategy to ensure your job postings are getting found.
How to make your job posting stand outGetting the attention of a candidate is hard enough, but once you get it, you need to ensure they go on to the next step, which is to apply for the job! This means that your job posting needs to be expertly crafted.
Here are some tips for how to make a job posting stand out.
Think like a job seekerAnd, let’s be real: Pay and benefits are what every job seeker is looking for and are good reasons why they should come work for you! Try to include these details if you can.
Related: 13 talent sourcing strategies to try.
Use clear language (no jargon!)Be concise and grammatically correct across all job post elements, including the title listed and job description. Avoid special characters, abbreviations, or in-house terminology such as:
Machinist II (Machinist for/of what? And if there is a II is there a I and III, too, and what does that mean?)
Manager 3rd Shift (Manager of what and when does the 3rd shift start?)
By avoiding jargon and making your job listings clear and to the point, you can encourage more applications.
Always start your ad with the job title, and keep it simple and straightforward such as:
CDL Drive Class A
Nurse LPN
Store Clerk
Starting an ad with the header Help Wanted or Hiring Now is not effective—no one searches for Help Wanted. People search for jobs based on skillset such as driver, nurse, store clerk, etc., or by job title.
Pick your location wiselyIf you’re trying to fill a position in a remote area, post your job ad to the largest city or town within a 25-mile radius. Sometimes posting the job to the county of the location may prove more effective.
If your job listing is for a remote work or work from home position, make sure you clearly state if this permanent or temporary, if in-person training is required, and if you provide any support or assistance to get your employees set up.
This job posting was listed as Remote, Full Time, and lays out additional benefits for WFH employees.
Also, note if they can remote work from anywhere or within 100 or more miles from the training center. Be as specific as possible.
Related: Get tips for managing remote employees.
Don’t be afraid to get creative!When composing your job posting, do not hesitate to insert your own brand and voice into the content you are writing. All things considered…Would you apply to this job? If the answer is no, it might be time to go back to the drawing board and add elements that make the position and the company sound exciting and appealing.
Play up what makes your company unique. Do you provide paid time off for volunteering? Do you offer employees a budget for learning or conference opportunities? Do you have a really cool mission statement? Your job posting is the place to brag about your business and why people love working there!
The intro in this job posting for Slack gives you a glimpse at their culture and their mission.
Start getting more applicants with job postings that stand outIf you find you’re not getting the volume of applications you expect, then you may not be visible enough online. Recruitment marketing can help you build awareness for your open positions and drive qualified candidates through your door. Learn more about the best recruitment marketing strategies here.
Anna BrekkaAnna has been interviewed by and quoted in The Florida Times-Union, NPR, Workforce Management, Chicago Tribune, HRExaminer, The Union Leader, and Recruiting Trends.
Other posts by Anna Brekka
3 Skills Every It Professional Needs To Stand Out
3 Skills Every IT Professional Needs To Stand Out
When you think of your job as an IT professional, what required skills usually come to mind? Most in the IT business would answer that an understanding of computers and their processes is essential for the job (and rightly so). But do soft skills such as listening, networking, and training come to mind? If not, read on to find out why they should. You might be surprised!
Listening
While most of your work will likely be relegated to computers, it’s important to remember that every project you work on will likely involve human interaction to some extent. This is where your listening skills come in handy.
Think of that one client that always has an issue with their network. While it would be easy to assume you know what the problem is based on what your client tells you, it’s vitally important to be able to pick up on what they’re not telling you. This can only be achieved by listening to what your client has to say. They may be describing the same issue to you (again), but this time they casually mention that their system hasn’t been working properly since a major thunderstorm passed through their area. This is just the clue you need to be able to follow up with the question, “Did you have a power outage during that storm?” Their system may need a quick reset because of the outage, and if you weren’t listening, you could’ve spent hours of your precious time (and your client’s billable hours) trying to fix a minor problem.
Networking
The social kind, that is. Networking is an often overlooked tool in the business world but, if done properly, could easily be the biggest tool in your arsenal. In fact, Business Network International (better known as BNI), estimates that the usual member in their networking organization has averaged $37,000 worth of business in the past 12 months, all due to networking. (It’s worth noting that BNI is a membership-based organization, but the message here is that networking of any kind is likely to get you more business than no networking at all.) If that’s not enough to get you to start networking, think about this: building a network of strong, reliable business contacts could save you from having to turn down a major job just because you don’t have the time, manpower, or expertise to complete it.
Picture it: you’ve just received a call from one of the larger hospitals in your area and the computers at the nurses station on the 2nd floor have all gone down and the hospital needs you right away. You know you have expertise in system interfaces, but it sounds like the problem may also be an electrical one. You may not have electrical skills, but if you’ve networked properly (i.e., created relationships with people who have IT-related skills that you lack), a quick call to one of your contacts could solve that problem. By bringing your contact into the job, they can offer you the benefit of their knowledge, they gain some revenue, and you gain points with the client because you didn’t leave them the chore of having to find an additional person to take of their secondary issue. That’s a win for everyone!
Ongoing Training
Although technically not a skill, it’s worth mentioning. As an IT pro, you know better than anyone that technology changes faster than most people can keep up with. It should be a requirement that you (as well as your entire staff, if you have any) keep up with technology as it comes along. Yes, everyone.
This includes the people who answer your phone, manage your social media, and answer your emails. Because they are often the gatekeeper between your business and its clients, they should also be kept in the loop about new technologies and how they fit into the services you offer. Moreover, the company should provide corporate sales training or negotiation training to those who have to work directly with clients on a daily basis.
Imagine for a moment that you trained all of your IT staff to handle the latest technology that came out and they are well-versed and able to accurately address any issues that come along for your clients. Imagine again that you failed to keep your office staff up-to-date on this training and a potential client calls wanting to know if you can fix the issue his company is having. If you are able to gain this person as a client, the job is worth $10,000. Since your office staff has never heard of that issue before, they sound unsure and timid on the phone, leaving the potential client questioning your abilities as an IT professional and reluctant to hire you. As you can see, the cost of losing a new client because of improperly-trained staff far exceeds the cost of training them.
While these skills aren’t the traditional skills you’d think of when it comes to being an IT professional, they are important nonetheless. If you employ just one of these skills in your business in the coming months, you’re sure to be able to do something for your clients that your competitors can’t: bring value to the money they spend.
How To Make Google Assistant Read Your Articles Out Loud
Found an interesting article you’d like to read but too tired right now or busy doing something else? You could save it for later or ask your Google Assistant to read it to you instead. This way you won’t risk forgetting about it afterward.
The feature was recently introduced. Google explained that the technology is different from other screen-reading software, as it’s capable of reading text in a natural-sounding voice and cadence so that people won’t have trouble listening and understanding, even for longer periods of time.
Google’s virtual assistant is available on most Android phones, so if you’d like to hear it read out to you, all you need to do is follow a few simple steps.
Note: so far the feature is meant for listening to articles, blog posts, or short stories online.
Set Up the Voice Match on Your Device (If It’s Not on Yet)To check if your Google Assistant is already active and ready on your device, touch and hold its Home button or say “OK Google” or “Hey Google.” If nothing happens, it means you’ll need to manually turn on Voice Match on your device. Here’s how to do it:
1. Go to your device’s Settings app.
2. Scroll down until you find Google and tap on it.
4. Find and tap on the “Voice Match” feature under the Hey Google category.
5. Make sure the Google Assistant toggle is on. Also, turn on “Hey Google” under Voice Match.
6. Once Hey Google is on, Google will ask you to say a few phrases so that it can learn your voice. Follow the steps indicated.
Note: on some older devices that don’t offer the assistant’s services by default, you may have to download the Google Assistant app.
Get Google Assistant to Read Articles to YouBefore we start, we need to point out one thing. The feature only works if you’re using the Chrome browser on your Android device or any other Chromium-based browsing app. If you have one on your phone, you are good to go.
1. Open an article you’d like to read in your browser of choice, then simply say: “Hey Google, read it.”
2. The first time you’ll attempt this command, Google Assistant will probably say it can’t do that because “Screen Context” needs to be turned on first.
3. A pop-up window will promptly appear on the screen asking you to give the Google app access to use your screen context. Tap OK. You don’t have to go looking through Settings on your own.
4. Now go back to the article and ask the Assistant once again to “Read It.” This time it will surely oblige and say “Got it” or “OK.” It takes a few seconds, as the virtual helper scans the piece before starting to read it out loud.
Take Full Advantage of the ControlsYou can access these controls directly from the Notification tray as well. You can even close the browser and still be able to pause or play the reading just by swiping down from the home screen or directly from the lock screen.
But wait – there’s more. Can’t understand English all that well? Perhaps you would like to hear the virtual assistant read to you in your native tongue.
Look for the three-dot menu icon located in the upper-right corner of the “Read It” browser and tap it, then on Translation. Here you can choose the language you want the text translated to, including German, Spanish, Polish, Romanian, and more (42 languages in total). The feature leverages the power of Google Translate to achieve this, and the results are remarkably adequate and convincing.
From the same menu users can select a different voice to read their articles. You’ll find this option under “Read aloud voice.”
Additional ConsiderationsAs we mentioned above, you’ll need a Chromium-based browser for this feature to work. What’s more, “Read It” is not available for certain websites and most apps. (You’ll need to use the web version instead.) If you’ve stumbled upon a paywall or a log-in is requested to display the full article, the feature may not work. Despite these limitations, the new Google Assistant trick proves super useful. Going forward, Google will surely continue to improve it.
Want to get even more out of your Google Assistant? You may want to learn how to create an emergency routine using Google’s virtual helper, just to stay on the safe side of things.
Image Credit: Google Blog
Alexandra Arici
Alexandra is passionate about mobile tech and can be often found fiddling with a smartphone from some obscure company. She kick-started her career in tech journalism in 2013, after working a few years as a middle-school teacher. Constantly driven by curiosity, Alexandra likes to know how things work and to share that knowledge with everyone.
Subscribe to our newsletter!
Our latest tutorials delivered straight to your inbox
Sign up for all newsletters.
By signing up, you agree to our Privacy Policy and European users agree to the data transfer policy. We will not share your data and you can unsubscribe at any time.
Tricks For Data Visualization Using Plotly Library
This article was published as a part of the Data Science Blogathon
Data is everywhere you just need an eye to select which data is useful, by keeping stories interesting. That doesn’t mean you have to only just show graph and work is done it is the role of the data visualizer how to present the right data which helps the business to grow and have a powerful impact.
DataThe Data Which we are going to use is available here and the description of the data is available here
Overview of Data:The data tell us which products are recommended on basis of Ratings, Reviews of products, and many other factors.
Clothing ID: Integer Categorical variable that refers to the specific piece being reviewed. Age: Age of the reviewer’s age. Title: The Title of the review. Review Text: The description of the product by customers. Rating: Ratings were given by the customer to a different product from worst 1 to best 5 Recommended IND: Binary variable stating where the customer recommends the product where 1 is recommended, 0 is not recommended. Division Name: Categorical name of the product high-level division. Department Name: Categorical name of the product department name. Class Name: Categorical name of the product class name. Positive Feedback Count: Positive Integer documenting the number of other customers who found this review positive.Original DataFrame looks Like:
Table of Content1. what is plotly
2. Points to keep in mind while designing graph
3. Data visualization graph configuration
Univariate visualization
Bivariate visualization
Multivariate visualization
4. Chart Types
Pie Chart
Histogram Chart
Stacked Histogram Chart
Box Chart
Funnel Chart
TreeMap Chart
HeatMap
Scatter Matrix
5. Embedding charts in a blog with Chart Studio
6. Plotly Dash
What is Plotly?Plotly is an open-source library that provides a list of chart types as well as tools with callbacks to make a dashboard. The charts which I have embedded here are all made in chart studio of plotly. It helps to embed charts easily anywhere you want.
The main plus point of plotly is its interactive nature and of course visual quality. Plotly is in great demand rather than other libraries like Matplotlib and Seaborn. Plotly provides a list of charts having animations in 1D, 2D, and 3D too for more details of charts check here.
If you just want to embed charts in your blogs you don’t need to have prior knowledge of coding or javascript you can just use chart studio, where you just need to select the parameters and your chart is ready.
If you want to make a dynamic dashboard, Plotly provides Dash which is a plotly extension for developing web applications. for more details check plotly documentation here.
Points to keep in mind while designing graph1) No need to keep all the data in one graph.
It is always better to divide and rule.
Always apply filters to your graphs to make them more interactive.
2) Sometimes displaying data in form of a card is also a great way of representing data.
As you see in the card layout we can use infographics to enhance the data.
As you see in the graph & card layout both show the same information but in different ways with the help of plotly library.
I will show you two charts tell me which helps you to understand better.
The graph shows how many people have given positive, negative, and neutral reviews for a product.
3) Styling the graph
The thing which I have observed is most of the time people overdue to it in different ways like they will put different styling in one graph only.
I will show you two charts one will be right and another one is to avoid.
As we are using dark background so title color should be eye-catching prefer light colors. In my case, I have used white which usually looks better with dark backgrounds.
Don’t use the different color labels for each category like in my example red for Asia, green for Europe.
Try to avoid different colors for each category as shown in the wrong graph where one category uses red other one uses green. The graph doesn’t look professional and looks too crowded. If possible use a sequential palette.
Always keep in mind that the color of the title and category label should be different for easily differentiable.
There are others things to keep in mind while designing graphs, which we will discuss in the later section.
Keeping in mind these simple steps that will help you to get your work easily done.
Data visualization graph configurationMainly, there are three types of analysis for Data Visualization:
Univariate Analysis: In the univariate analysis, We will use a single feature to visualize
Bivariate Analysis: In the bivariate analysis, We will compare two features for visualizing.
Multivariate Analysis: In the multivariate analysis, We will compare more than two features for visualizing.
Let’s start how to use Plotly for making graphs.
Installation
Install with pip or conda
# pip pip install plotly # anaconda conda install -c anaconda plotlyWhile importing the plot you should install the pandas library first otherwise there will be an error.
#Importing library import plotly.express as px fig.update_layout(layout_parameters or add annotations) fig.update_traces(further graph parameters) fig.update_xaxis() # or update_yaxis fig.show()Using update_traces we can change the text font color, size
Using update_layout we can add graph parameters. Below I have explained every parameter.
Chart Types: 1. Pie chart
The pie chart is mostly used for categorical data when you have more than 2 categories it is easy to compare.
division_rat = px.pie(df, names='Rating', values='Rating', hole=0.6, title='Overall Ratings of Products', color_discrete_sequence=px.colors.qualitative.T10) division_rat.update_traces(textfont=dict(color='#fff')) division_rat.update_layout(autosize=True, height=200, width=800, margin=dict(t=80, b=30, l=70, r=40), plot_bgcolor='#2d3035', paper_bgcolor='#2d3035', title_font=dict(size=25, color='#a5a7ab', family="Muli, sans-serif"), font=dict(color='#8a8d93'), legend=dict(orientation="h", yanchor="bottom", y=1.02, xanchor="right", x=1) )
Interpret:
As we see in the graph 5-star ratings are 66% given to the products so overall products are nice.
2. Histogram ChartFrom a histogram, we can see how one category differs from the other like which is highest and lowest.
classname2 = px.histogram(df, x=’Department Name’, title=’Recommended IND by Class Name’, height=250, color_discrete_sequence=[‘#03DAC5′], ) classname2.update_yaxes(showgrid=False), classname2.update_xaxes(categoryorder=’total descending’) classname2.update_traces(hovertemplate=None) classname2.update_layout(margin=dict(t=100, b=0, l=70, r=40), hovermode=”x unified”, xaxis_tickangle=360, xaxis_title=’ ‘, yaxis_title=” “, plot_bgcolor=’#2d3035′, paper_bgcolor=’#2d3035′, title_font=dict(size=25, color=’#a5a7ab’, family=”Muli, sans-serif”), font=dict(color=’#8a8d93′), legend=dict(orientation=”h”, yanchor=”bottom”, y=1.02, xanchor=”right”, x=1) )
Interpret:
Here as we see tops are generally more preferred compared to jackets
3. Stacked Histogram chartFrom a stacked histogram we can easily compare two quantities against each other.
classname = px.histogram(df, x=’Department Name’, color=’Recommended IND’, title=’Recommended IND by Class Name’, height=300, category_orders={‘Recommended IND’: [‘Recommended’, ‘Not Recommended’]}, color_discrete_sequence=[‘#DB6574’, ‘#03DAC5′], ) classname.update_yaxes(showgrid=False), classname.update_xaxes(categoryorder=’total descending’) classname.update_traces(hovertemplate=None) classname.update_layout(margin=dict(t=100, b=0, l=70, r=40), hovermode=”x unified”, xaxis_tickangle=360, xaxis_title=’ ‘, yaxis_title=” “, plot_bgcolor=’#2d3035′, paper_bgcolor=’#2d3035′, title_font=dict(size=25, color=’#a5a7ab’, family=”Muli, sans-serif”), font=dict(color=’#8a8d93′), legend=dict(orientation=”h”, yanchor=”bottom”, y=1.02, xanchor=”right”, x=1) )
Interpret:
Most of the products are recommended and the ratio of recommended to non-recommended products is too much, which is a great sign.
4. Box plotBox plot is a great option whenever we want to look for the outliers. It will give the range where most of the data lie in quartile ranges.
fig_box = px.box(df, x='Age', title='Distribution of Age', height=250, color_discrete_sequence=['#03DAC5'], ) fig_box.update_xaxes(showgrid=False), fig_box.update_layout(margin=dict(t=100, b=0, l=70, r=40), xaxis_tickangle=360, xaxis_title=' ', yaxis_title=" ", plot_bgcolor='#2d3035', paper_bgcolor='#2d3035', title_font=dict(size=25, color='#a5a7ab', family="Muli, sans-serif"), font=dict(color='#8a8d93'), legend=dict(orientation="h", yanchor="bottom", y=1.02, xanchor="right", x=1) ) 5. Funnel chartA funnel chart is mainly used when we have it in a decreasing manner like in sales data or company size.
df_rec = df[df[‘Recommended IND’] == ‘Recommended’][[‘Recommended IND’, ‘Department Name’]] df_rec_dep = df_rec[‘Department Name’].value_counts().rename_axis(‘Stage’).reset_index(name=’Counts’) df_rec_dep[‘Recommended IND’] = ‘Recommended’ df_not_rec = df[df[‘Recommended IND’] == ‘Not Recommended’][[‘Recommended IND’, ‘Department Name’]] df_not_rec_dep = df_not_rec[‘Department Name’].value_counts().rename_axis(‘Stage’).reset_index(name=’Counts’) df_not_rec_dep[‘Recommended IND’] = ‘Not Recommended’ dff = pd.concat([df_rec_dep, df_not_rec_dep], axis=0) department = px.funnel(dff, x=’Counts’, y=’Stage’, color=’Recommended IND’, height=300, title=’Recommended IND by department Name’, category_orders={‘Recommended IND’: [‘Recommended’, ‘Not Recommended’]}, color_discrete_sequence=[‘#DB6574’, ‘#03DAC5′], ) department.update_traces(textposition=’auto’, textfont=dict(color=’#fff’)) department.update_layout(autosize=True, margin=dict(t=110, b=50, l=70, r=40), xaxis_title=’ ‘, yaxis_title=” “, plot_bgcolor=’#2d3035′, paper_bgcolor=’#2d3035′, title_font=dict(size=25, color=’#a5a7ab’, family=”Muli, sans-serif”), font=dict(color=’#8a8d93′), legend=dict(orientation=”h”, yanchor=”bottom”, y=1.02, xanchor=”right”, x=1) )
Interpret:
The Tops is the highest product which is recommended by 7047 peoples.
Funnel chart always helps to show the data in decreasing fashion
6. TreeMap fig = px.treemap(df, path=[px.Constant("Tree Map"), 'Division Name', 'Department Name'], color_discrete_sequence=['#DB6574', '#03DAC5', '#0384da'], values='Rating') fig.update_layout(margin = dict(t=50, l=25, r=25, b=25), height=300, plot_bgcolor='#2d3035', paper_bgcolor='#2d3035', title_font=dict(size=25, color='#a5a7ab', family="Muli, sans-serif"), font=dict(color='#8a8d93'))
People usually recommended General division products than General Petite and last is intimates Products.
In General Division Most of the people recommended Tops than Dresses.
7. HeatMapWhenever we need to see the correlation between the data it is always the best option to go with heatmap.
import plotly.figure_factory as ff # Heatmap # Correlation between the feature show with the help of visualisation corrs = dff.corr() fig_heatmap = ff.create_annotated_heatmap( z=corrs.values, x=list(corrs.columns), y=list(corrs.index), annotation_text=corrs.round(2).values, showscale=True) fig_heatmap.update_layout(title= 'Correlation of whole Data', plot_bgcolor='#2d3035', paper_bgcolor='#2d3035', title_font=dict(size=25, color='#a5a7ab', family="Muli, sans-serif"), font=dict(color='#8a8d93')) 8. PairplotPairplot is mostly used when we need to find the relation between different categories.
dff = df[['Age', 'Rating', 'Recommended IND', 'Class Name']] fig_pairplot = px.scatter_matrix(dff, height=500, color='Recommended IND', title= 'Correlation of whole Data') fig_pairplotInterpret:
As we see there is a positive relation between Age and Recommended IND.
1-star, 2-star rating products are not generally recommended.
Embedding charts in a blog with Chart StudioInstalling chart studio
# pip pip install chart_studioSetting the chart studio
import chart_studio import chart_studio.plotly as py import chart_studio.tools as tls chart_studio.tools.set_credentials_file(username=' ', api_key=' ')2. Installing the library run any code which is present above for example run a pie chart
3. Run the below code
py.plot(figure_name, fielname='Pie chart', auto_open=True)After completing all the 3 procedure chart studio will open scroll down you will see the embed option just copy-paste the link and the graph is embedded.
Plotly DashIf you want to make a dynamic dashboard, Plotyy provides Dash which is a plotly extension for developing web applications. for more details check plotly documentation here.
To make the dashboard looks good plotly provides Css, Html, Bootsrap, react too.
The media shown in this article are not owned by Analytics Vidhya and are used at the Author’s discretion.
Related
Future Scope & Key Trends In Data Visualization
From prehistoric cave paintings to the pictogram-based writing systems of the ancient Babylonians and Egyptians, the affinity for codifying information as visuals has been a defining, persistent characteristic of human behavior.
Many contemporary world languages are still heavily pictogram-based, with modern Chinese serving as the most long-lived example. And as human languages are in continuous evolution, so are the means and methods for encapsulating knowledge in images.
In the context of the information age, this human trait is manifest in the design of visually compelling, highly interactive experiences for sight-based faculties of human perception. By providing digital visual representations of physical objects, data visualization enables human operators to more easily manage vast data sets, glean insights from a myriad of information sources simultaneously, and perform powerful operations more intuitively and tactilely.
Data visualization can enhance the value of information by incorporating motifs, objects, and imagery native to a specific use case and/or industry.
From agriculture data visualization used in prescriptive crop planning to augmented reality (AR) in financial services for mapping out data-driven wealth management scenarios, industry-specific applications of data visualization technology are enabling businesses and consumers alike to make better-informed decisions.
Indeed, global market demand reflects the growing pervasiveness of data visualization. In 2023, data visualization was valued at $9.06 billion and is projected to grow at a compound annual growth rate (CAGR) of 7.83% for a market size of $15.35 billion by 2026.
See more: How Data Visualization is Taking Form for Home Depot, Emblem Health, Singapore, Members First Credit Union, and Geospatial
The following trends in data visualization reflect the general move toward use-case optimized visual experiences and the accessibility of data visualization across both devices and industries.
Digital twins are virtual models of physical objects/systems created by pulling in data streams related to the physical asset in question (e.g., telemetry from onboard sensors monitoring temperature, vibration level, etc.). This enables the remote monitoring of performance and health/condition parameters, allowing for physical assets to be analyzed and assessed from afar.
In the past, these digital models were presented to users in the form of interactive dashboards and continuously updated metrics. Newer offerings such as Oracle’s IoT Asset Monitoring Platform and Microsoft Azure Digital Twins integrate data streams with 3D asset models for truly high-fidelity digital twins — the ultimate in data visualization.
With software as a service (SaaS) being today’s preferred way to consume software, web front ends are the primary interfaces between applications and users. In this space, technologies like Flash and Java have all but died out, while JavaScript continues to reign supreme. These days, popular JavaScript frameworks such as chúng tôi chúng tôi and chúng tôi are used to streamline the development of complex front-end visualizations, while specialized frameworks like chúng tôi and chúng tôi add 3D and immersive reality to JavaScript-based data visualization.
As traditional industries undergo digitization, data visualization will become more specialized to the needs of specific industry audiences. For example, data visualization in shipping and maritime is enabling ship owner/operators to improve vessel performance and monitor safety and operational conditions. Similarly, the automotive industry is using data visualization to optimize vehicle product development workflows.
Data analysis and management systems were some of the first applications to incorporate artificial intelligence (AI)/machine learning (ML) for automating information collection, analysis, and dissemination. Similar trends can now be observed in the data visualization space, with automated systems leveraging ML models trained on common user patterns and task execution to construct UI dashboards. These components are automatically fine-tuned for delivering relevant, unique visualizations and insights per user. In the future, software solutions will increasingly rely on AI/ML for optimizing data visualizations used in human-computer interactions (HCI) .
See more: Top Data Visualization Tools for 2023
Update the detailed information about Unique Data Visualization Techniques To Make Your Plots Stand Out on the Daihoichemgio.com website. We hope the article's content will meet your needs, and we will regularly update the information to provide you with the fastest and most accurate information. Have a great day!