LMAC Image Library statistics - What is much | what is rare

in Let's Make a Collage3 years ago (edited)

Cover image

Dear Community,

The image business is booming at LMAC. Since we started the LMAC Image Library initiative, every day there have been dozens of new images contributed to the LIL. The initiative has been running for a few weeks now, so it was about time to check whether our child is growing up healthy.
I have made the effort to do a little keyword analysis of all LIL entries to be able to determine what we have a lot of and what kind of images are very rare.
The data obtained was very revealing, which is why I would like to share it with you, so that together we have the opportunity to fill thematic gaps in the LIL and more.
But also about the number of contributions per author I was able to determine data, which I would also like to share with you very much, simply to honour all those who have been particularly diligent in helping to build the content of the LIL. However, we are also a competitive community, which is why I have limited it to a top 10. :-D

Author statistics

A truly huge number of 66 Hivers donated all the 4112 images to the library. I proudly present the Top 10 of these authors.
(Based on the contributions that could be successfully read by the LIL bot so far.)

Keyword statistics

I have made list of most often used keywords, sorted by frequency of use in LIL contributions. For this, I took the painful trouble to match keywords from various websites on the internet, the LMAC community and a SEO keyword list with the keywords we have in the database to get the data of the most popular search keywords.
With this data we're wonderfully able to deduce which kind of images are still very rare in the library and which are very often to find there.

But, no guarantee that this keyword constellation is 100% correct. Because I had to make a part of the alignment with various sources (galleries and the like) by using a kind of a little web crawler program I developed myself. Difficult to validate that with one hundred percent certainty in view of that huge wild bunch of data out there and the different circumstances under which my program has found them.
However, some of the keywords below are obviously keywords that are quite popular, even in our community.

These are the keywords that my program found in the in some of the relevant places of the WWW as a basis for the keyword analysis:
Basically, most of them are the keywords found in the image entries of image top lists. So these are the keywords of the most often found images on the different websites, the successful ones.
As you will surely notice, many terms are in plural. Using plural was partly easier for my analysis.

animals, arts, humans, flowers, trees, cars, nature, sky, clouds, sea, landscape,
beach, mountains, hills, sun, moon, stars, balls, weapons, fruits, bicycles, 
gates, houses, streets, stones, fences, walls, devices, rivers, fungus, rocks,
spring, fall, autumn, winter, summer, castles, night, day, creatures, coffees,
dinosaurs, fantasy, food, jard, vegetation, water, bottles, leaves, boat, 
airplane, horizon, sunset, palms, islands, city, environment, herbs, fishes,
weather, travel, valley, old, antique, cutouts, textures, lights, lanterns, lamps,
insects, spiders, hiking, feathers, buildings, music, paintings, weeds, 
transparent, drawing, decoration, marine, glasses, metals, woods, mushrooms,
earth, planets, horror, alien, gold, silver, diamond, atmosphere, life, figures,
ornaments, papers, robotics, mechanics, sports, funny, persons, preys, 
predators, trucks, outdoors, explosions, minerals, eyes, skeleton, children, 
evil, good, umbrellas, vehicles, hats, honeycombs, hive, lightnings, dragons,
technology, organics, history, restaurants, furnitures, monuments, bonfires, 
fairy, boxes, agricultures, stairways, farming, sounds, machines, tunnels, 
ladders, hotels, roads, people, skulls, beefs, panels

The corresponding keywords used in the LIL contributions:

Syntax: keyword(frequency of use N)

nature(352), sky(314), animal(258), 
flower(202), beach(192), sea(186), clouds(169), 
mar(132), flowers(131), tree(130), vegetation(129), 
water(128), insect(121), leaves(117), landscape(117), 
trees(113), mountains(96), transparent(94), red(92), 
art(92), stone(86), rocks(77), birds(71), 
food(70), jard(68), hills(68), fantasy(67), 
rock(66), ave(66), insects(58), mountain(58), 
dog(57), cat(54), house(49), boat(47), 
drawing(45), river(39), light(37), bird(36), 
sun(36), wood(36), fruit(35), car(34), 
horizon(34), building(34), sunset(30), decoration(30), 
land(27), palm(23), cloud(23), island(23), 
marine(23), herbs(23), street(22), stones(22), 
fun(22), city(21), wall(21), glass(21), 
environment(21), road(20), hiking(20), antique(19), 
metal(19), lamp(18), fish(18), old(18), 
the(18), predator(18), hill(17), summer(17), 
monument(16), weather(16), travel(16), valley(16), 
mushroom(16), name(16), earth(15), cutout(15), 
outdoor(14), dogs(14), minerals(14), day(13), 
fire(13), texture(13), lantern(13), gold(13), 
horror(13), alien(12), autumn(12), spider(12), 
atmosphere(12), buildings(11), feathers(11), music(11), 
ball(11), weed(11), life(11), airplane(11), 
painting(11), figure(11), fence(10), ornament(10), 
bottle(10), weeds(10), glasses(10), 
paper(10), bee(10), mechanics(10), robotics(10), 
spare(10), room(9), person(9), fruits(9), 
prey(9), spiders(9), 
cars(8), sound(8), truck(8), air(8), 
paint(8), isla(8), outdoors(8), houses(7), 
explosion(7), man(7), dinosaurs(7), coffee(7), 
creature(7), moon(7), night(7), wing(7), 
spring(7), castle(7), gate(7), honey(7), 
mineral(7), fungus(7), eyes(6), door(6), 
skeleton(6), horse(6), bicycle(6), monster(6), 
children(6), aves(6), child(6), evil(6), 
and(6), weapon(6), ladder(5), umbrella(5), 
funny(5), vehicle(5), hat(5), honeycomb(5), 
hive(5), star(5), ornaments(5), dragon(5), 
technology(4), ring(4), eye(4), organic(4), 
hot(4), decor(4), sport(4), fall(4), 
silver(4), history(4), figures(4), restaurant(4), 
furniture(3), sphere(3), monuments(3), sports(3), 
all(3), bonfire(3), people(3), fairy(3), 
cows(3), herb(3), box(3), cats(3), 
agriculture(3), streets(3), stairway(3), 
dia(3), las(3), port(3), ant(3), 
culture(3), iron(3), farming(3), hotels(3), 
net(3), eco(3), mount(2), sounds(2), 
machine(2), tunnel(2), hotel(2), roads(2), 
bottles(2), log(2), walls(2), palms(2), 
stars(2), way(2), rest(2), device(2), 
dinosaur(2), lane(2), rat(2), panel(2), 
skulls(2), skull(2)

While browsing the keywords, you have probably already noticed which image categories are contributed to most frequently and to which not that often. Even within the different top categories, the frequency of images in individual subcategories is extremely out of balance.

Example (Top category->subcategory):

Top category: plant
- Sub category: flowers
- Sub category: trees
- Sub category: herbs
...

For example, let's look at the top category "animal", which contains 258 images in total. Although "animal" was not used as a keyword for all animal species, in view of these numbers, the imbalance can be seen very clearly here.
"insects"/"insect" make up the majority of all animal images with 179 images. We have 107 images tagged with "birds"/"bird".
However, the keywords "dog" and "dogs" together have only 71, "cats" and "cat" have only 57 images, although both types of animals are actually very popular and very often used in artworks.
Unfortunately there are only 3 images in the LIL which have a "cow" or "cows" keyword, only 6 horses...

Friends, we need more species in the library. Let's go out and take pictures of all kind of animals. The neighbor's dog will be happy when you visit it to take some photos. Dear Americans, you have cowboys on horses herding cows... In big cities crocodiles are said to live in the sewers, just as a tip! Exciting! :-D
In any case, it would be really great if we were able to balance it out.

Another example of what could be improved is the keyword list below. These are the top 30 non-aligned keywords.

cielo(408), nature(352), naturaleza(326), 
sky(314), green(293), animal(258), flower(202), 
arboles(195), beach(192), sea(186), garden(182), 
clouds(169), nubes(169), flor(142), yellow(141), 
plantas(140), playa(136), monta(134), mar(132), 
flowers(131), tree(130), vegetation(129), water(128), 
blue(125), plant(121), insect(121), leaves(117), 
landscape(117), flores(117), planta(116) 

As you can see, there are actually huge categories like "cielo" whose images can only be found by Spanish speakers. So I have to develop soon a program that translates the non-English keywords to include them additionally, so that they can be found by more people. This will be very heavy work for the server, that's why I wouldn't want to run the program permanently.
I would therefore like to encourage you to always add at least additional English keywords to the images to be contributed. You can earn more from your images when more people find them. This is in fact a little bit a kind of a SEO (Search Engine Optimization) task.

Rare types of images

... often used in collages.

  • creatures
  • cutouts
  • eyes
  • lanterns
  • aliens
  • dragons
  • fishes
  • cars
  • stars
  • castles
  • moon
  • sun

And we need much more animal species of all kinds, compared to the insects(179) and the birds(107).
We also need more images of other plants species, compared to flowers(333).

It would be so great if we could fix these unbalances together. :-)

How to improve your posts to be found better

1. Pluralized keywords

Most nouns simply get an "s" suffix to write them in plural form. That's fantastic, because so it is very easy to make an image findable for both, singular and plural keyword nouns by just writting the keyword in plural only.
The searcher might not be too annoyed of a search result with photos of two or more bicycles in case of having entered the singular term "bicycle".
If it is not possible to pluralize a specific keyword by adding a suffix, you could just use the plural in addition.

pluralkeywords.jpg

2. Variance

The more variants you use in your keywording, the higher the chance for your images to appear in search queries.
People think differently, even when it comes to entering terms in search queries. Culture, education, gender, social background etc... are all factors that influence this. Therefore, it can have a positive effect on the findability of your images if you keep this in mind when doing your keywording.
To search for a photo of a teddy bear or the like, many european women also would enter "cuddly animal" while many men would most likely enter "soft toy". Not to forget that US-Americans might even type in "stuffed animal", I guess.

More examples...

Example 1.:
AU: Hot chips
UK: Chips
US: French Fries

Example 2.:
UK: Organisers
US: Organizers

Example 3.:
UK: Trousers
US: Pants

Example 4.:
Elegant: Automobiles
Casual: Vehicles
Common: Cars

Example 5.: 
Common: Sunglasses
Casual: Shades

For a photo showing pants, it obviously is a good idea to use "pants" as well as the keyword "trousers".
However, some people might not necessarily enter "trousers" or "pants" in the search field because they assume that they will get more results with a more general term.
So let's just expand the keyword list of our pant photo: "trousers", "pants", "textiles", "clothes", " fabrics". Now, it can be found in many cases.


Well, what I'm pointing out is... Let's use the data above to improve the diversity and the efficiency of the library together, by improving our keywording whenever we contribute to LIL and by trying to contribute a little more what is too rare. We will find easier, faster and much more in the gallery as a result and those who contribute have a good chance to earn much more. :-)

Best regards
QuantumG


╭━━⋞ ☙ My NFT artworks ≻≺ ♖ My dCity ⋟━━╮
╰━━━━⋞ ♫ My Rising Star(s) ⋟━━━━╯

Sort:  

Great analysis, thank you @quantumg. Also, I notice these days, when I'm looking for Halloween objects, we need more png objects with transparency. I'll do something in my new #lil post.

Awesome!!! I'm very glad that you help us to get out of this dilemma and am very curious about your next LIL post. :-)

We are so proud of you, @quantumg, and amazed. We were lucky the day you joined LMAC. Thank you for helping us to grow and making an idea into a functioning reality. You opened the door to the future.

Great for LMAC. Great for Hive. Great for anyone looking for an opportunity to become part of a blockchain.

Wow! That's high praise indeed — thank you so much for such kind and appreciative words.
Shaka is a visionary administrator. I'm absolutely sure LMAC would have grown in a great direction regardless if I hadn't been around. Fate has given me the opportunity to be involved.
Let's look at it this way, I'm a nobody here in the Hive, don't have a huge Stake, am not one of the Top Hivers, ... So I am really just the lucky guy. :-D
Sometimes I can hardly believe it myself.

Thank you very much for the tip! :-)

We are all lucky that we stumbled on LMAC. I'm an early recruit. One of the things that held me was the apparent ethics of the leadership. Really important to me. Of course @shaka is a brilliant administrator, even a bit of a visionary. But if he didn't run this community with the ethics that are obvious, I wouldn't be here. And so, he attracts people like you. Isn't that natural?

This is very important information to maintain balance in sharing images in the LIL library, I think everyone should read the posts you show..
Let me reblog your post so everyone knows what's needed in the LIL library right now.

Thank you for helping to spread the message. :-)

Indeed the keywords solved is very important information to gain and maintain balance in the shared images in the LIL library, I think everyone should read the post and add what they have in their library. I would love a list of contributors aswell that is great to have embedded on the Blockchain

Loading...

Exclente información,en lo posible debemos tratar de seguir las sugerencias del amigo @quantum,para mejorar mucho mas la Blblioteca LMAC.LIL,Gracias feliz día

Thank you for keeping us informed

Your content has been voted as a part of Encouragement program. Keep up the good work!

Use Ecency daily to boost your growth on platform!

Support Ecency
Vote for Proposal
Delegate HP and earn more

Congratulations @quantumg! You have completed the following achievement on the Hive blockchain and have been rewarded with new badge(s) :

You got more than 1500 replies.
Your next target is to reach 1750 replies.

You can view your badges on your board and compare yourself to others in the Ranking
If you no longer want to receive notifications, reply to this comment with the word STOP

Check out the last post from @hivebuzz:

Hive Power Up Day - November 1st 2021 - Hive Power Delegation
Bee ready for the 2nd Hive Power Up Month challenge!
Support the HiveBuzz project. Vote for our proposal!

👏 @quantumg what a great job you have done. With this valuable information I am sure we will improve our keywording for our future contributions to LIL.
Thanks...

I am very happy to see this good information, where many things are still a homework for writers and photographers, in sharing interesting pictures that still have many opportunities for the LMAC Library. I ask permission to re-share this post. So that more people know this information.

Spreading that message is very welcome. Thank you :-)

Congratulations @quantumg! You received a personal badge!

You powered-up at least 10 HIVE on Hive Power Up Day!
Wait until the end of Power Up Day to find out the size of your Power-Bee.
May the Hive Power be with you!

You can view your badges on your board and compare yourself to others in the Ranking

Check out the last post from @hivebuzz:

Hive Power Up Day - November 1st 2021 - Hive Power Delegation
Support the HiveBuzz project. Vote for our proposal!

These are very timely tips, if we teach excellence that is what will multiply. And I think the use of keywords is very important, it helps to optimize the search and direct you to the site of your interest.

Greetings this is an excellent job thanks for contributing with the programming of the image library

Hello @quantumg. Good information on the keywords.

I see that I'm listed as having contributed only 110 images:
|@justclickindiva|110|Click|

Couple questions.

  • Did the bot have trouble reading my recent posts?
  • Do I need to return and revise them in some way?
  • I used the keyword "digital art". Should all keywords not contain any spaces?

I will try to locate additional images in the rare categories.

Thank you for the report on how to make the library much more functional and varied.

Hi,

I found one of your LIL posts in our list for manual review. When I checked it, it did not appear to contain any typos. Everything seemed to be correct. So I ordered the bot to read it again. This will be done in 2 days.

You can use this type of keywords. However, currently they are interpreted as two keywords. In one of the next software update cycles, they will be handled as one. So no further steps are required from your side. :-)

Thank you for taking the time to response and explain. I appreciate it as I know you're busy. Have a good rest of your week.

@agmoore -- respect! a huuuge respect to see you topping this list. hehe. 😎 but remember that quantity and quality are very closely connected together

Excelente post, muy interesante lo explayado, gracias por compartir, te sigo, saludos.

Gracias, @quantumg. Estaré atenta para mis próximas contribuciones.

Thanks information it sir @quantum

execelente post

A little tip: Have a look at "stemming algorithms". You can use it to find the root of a word and solve problems like using the plural/singular. For example, "cats" would become "cat" or "swimming" would turn into "swim"!

Hello @quantumg I like this post because it has given me better ideas of what I should do to display the images I have to contribute.

Thank you very much for dedicating time to all of us at #lmac.

A big hug.