=Paper=
{{Paper
|id=Vol-1520/paper9
|storemode=property
|title=Automated Blend Naming Based on Human Creativity Examples
|pdfUrl=https://ceur-ws.org/Vol-1520/paper9.pdf
|volume=Vol-1520
|dblpUrl=https://dblp.org/rec/conf/iccbr/PollakMCU15
}}
==Automated Blend Naming Based on Human Creativity Examples==
https://ceur-ws.org/Vol-1520/paper9.pdf
93
Automated blend naming based on human
creativity examples
Senja Pollak1 , Pedro Martins2 , Amı́lcar Cardoso2 , and Tanja Urbančič13
1
Jožef Stefan Institute, Ljubljana, Slovenia
2
CISUC, DEI, University of Coimbra, Coimbra, Portugal
3
University of Nova Gorica, Nova Gorica, Slovenia
senja.pollak@ijs.si, pjmm@dei.uc.pt, amilcar@dei.uc.pt
tanja.urbancic@ung.si
Abstract. In this paper we investigate which principles people use when
they name new things as results of blending. The aim is to uncover
patterns with high creative potential and to use them for automated
generation of names for new creations or phenomena. We collected ex-
amples with a web survey in which participants were asked to evaluate
pictures of animals with blended anatomies from two different animals,
and to provide their own names for blended creatures on the pictures.
The blended animals served as a trigger of human creativity manifested
through imaginative, humorous, surprising names collected in the survey.
We studied how the features from the pictures reflected in the names,
what are different complexity levels of lexical blend formation and how
far in other realms subjects “travelled” to search for associations and
metaphors used in the names. We used the findings to guide automated
generation of names for the blends.
Keywords: Computational creativity, human creativity examples, conceptual
blending, lexical blend generation, creative naming, bisociation.
1 Introduction
Creativity is in the core of many human activities and has been studied for
decades [9][2]. As a phenomenon challenging for being replicated with machines,
it became also a topic of artificial intelligence research [21]. While creativity is
an intriguing research question by itself, it is also a driving force of development
and as such, it has an immense value for applications in countless areas, includ-
ing scientific discovery, engineering inventions and design. One of the cognitive
principles underlying such discoveries and inventions is conceptual blending [5]
in which two mental spaces integrate into a new one, called blend. Conceptual
blending has also been implemented and tested in computer systems to produce
novel concepts [17]. However, there are still many open questions related to the
choice of input mental spaces and the ways of projections that lead to blends,
perceived as creative and inspiring. In our work we aim at providing guidance
Copyright © 2015 for this paper by its authors. Copying permitted for private and
academic purposes. In Proceedings of the ICCBR 2015 Workshops. Frankfurt, Germany.
94
for choosing input spaces and projections based on concrete findings about hu-
man creativity with elements of blending. More precisely, by investigating the
patterns that we can find in the cases of human creations, we guide the blending
process to the extent allowing for automated generation of blends.
Conceptual blending and case-based reasoning [10] can meet in a very fruitful
way in areas such as design and architecture [4][6].In such domains, blends are not
only a source of surprise, artistic satisfaction or inspiration, but have also their
own functionality, bringing into the process additional constraints and priorities.
Contexts and goals can also be used in computational approaches to conceptual
blending and can beneficially affect the issues of efficiency [13]. Authors in [1]
exploit a principle of creative transfer from one domain to another in the realm of
design. Their IDEAL system abstracted patterns from design cases in one domain
and applied them to design problems in another domain. connecting distant, self-
consistent and usually not connected frames of reference has been recognised and
used as an effective principle in the act of creation. Such connections of habitually
incompatible domains through common patterns or bridging concepts are also
referred to as bisociations [9].
In this paper, we address the issue of case-based reasoning and conceptual
blending in the context of lexical creativity. While this might appear quite far
from the discussion on design in the previous paragraph, the connection becomes
evident based on an observation by Veale and Butnariu [20]: “Words are every-
day things, as central to our daily lives as the clothes we wear, the tools we use
and the vehicles we drive. As man-made objects, words and phrases are subject
to many of the same design principles as the consumer artefacts that compete for
our attention in the market-place.”. The authors also draw attention to two basic
principles of artefact design, as identified in [15], namely visibility and mapping.
In the case of a well-designed product, the design should suggest a mental visual-
isation of a conceptually correct model of the product, and the mapping between
appearance and function should be clear. Their Zeitgeist system [20] can auto-
matically recognise neologisms produced as lexical blends, together with their
semantic meaning. This is done based on seven different “design patterns” recog-
nised in constructing neologisms as lexical blends. Types of lexical blends and
how new lexical blends are formed is described and illustrated with many ex-
amples in [12]. An important issue of recognising and quantifying creativity in
different combinations of words is studied in [11].
In our work we investigate how humans approach the task of naming new
things, and how based on human examples, a computer system could exhibit
similar (and, why not, better) performance. We consider this principle of using
past examples for revealing patterns to be used for new cases as a manifestation
of case-based reasoning. The concrete task was to name creatures – animals with
blended anatomies from two different animals. This was done in a web-based sur-
vey, designed primarily for a study of human perception of visual blends [14].
In this paper we continue using the material of the same study, but we exam-
ine it from a completely different angle, i.e. from the lexical creativity side by
investigating creative naming of blends. Many offers for supporting naming of
95
snorse chimporse durse
(snake, horse) (chimpanzee, horse) (duck, horse)
guineabear hammerheadhorse pengwhale
(guinea pig, bear) (hammerhead shark, horse) (penguin, whale)
proboscis parrot chamelephant duckphant
(proboscis monkey, bird) (elephant,chameleon) (elephant, duck)
guinea lion horbit hammerhead gull
(guinea pig, lion) (horse, rabbit) (hammerhead shark, gull)
horduck spider pig shark retriever
(horse, duck) (spider, guinea pig) (shark, labrador retriever)
Fig. 1. Hybrid animals dataset used in the online questionnaire (available at
http://animals.janez.me). Each sub-caption contains a name of the blend proposed by
survey participants, as well as the input spaces. All blends were created by Arne Olav,
with the exception of shark retriever and camalephant, whose authorship is unknown.
For a better visualisation, some images were slightly cropped.
96
client’s enterprises, products, etc. can be found on the web and show the appli-
cation potential of creative naming. The task has already been approached with
the goal of (semi-)automatic name generation and the results presented in [16]
and [18] demonstrate a very big potential. While our work shares some of the
ideas with above-mentioned related approaches, it differs from them by using
visually triggered human examples as examples used for automatic lexical blend
generation, and by using a novel categorisation of creativity level that guides
construction of blends based on bisociation as one of the key principles inherent
in many human creative processes.
After presenting the survey in which the names were collected in Section
2, we analyse different patterns and mechanisms used by people when coining
names in form of lexical blends in Section 3. These patterns are used in Section 4
for automatically generating blends of different levels. In Section 5 we discuss
the potential of our prototype and present further research perspectives.
2 Survey: Visual blends and their lexical counterparts
In [14], we introduced a survey consisting of an on-line questionnaire related to
the quality of visual blends. Around 100 participants assessed 15 hybrid animals
which were the result of blending anatomies from two different animals (Fig-
ure 1).The participants were asked to to rate criteria related to the coherence of
blends as well as creativity.
Clearly in our questionnaire on animal blends the main focus was on visual
blends. However, with the aim of getting more insight into potential connections,
participants were also asked to provide a name (in English, Portuguese, Slovene,
French or Spanish) to each of the hybrid creatures. By asking people to name the
creatures we wanted to investigate the following questions: Would participants
give names for all, for none, or for some of the creatures? How creative are they
when naming the animals, how does the visual blended structure reflect in the
lexical blend? Where the names provided by subjects mostly lexical blends or
not? Do lexical blends use animal’s “prototype” characteristics, or more sophis-
ticated associations for which some background knowledge is needed (like titles
of books, movies, history, etc.)? Does complexity of visual blends reflect in the
names? The names given to the visual blends are the focus of our study.
In our survey we collected 1130 names for 15 animals. The general trend was
that people gave more names at the beginning of the study and the trend of the
number of given names was descending. However, some pictures triggered more
generated names than expected by their position (e.g., guinea lion and spider
pig). The guinea lion is also the blend for which the unpacking (recognising
the input spaces) was the most difficult [14] and the one for which the highest
number of very creative, bisociative lexical blends were formulated.
97
3 Formation and complexity of lexical blends
Our previous investigations of relationship between conceptual blending and
bisociation have drawn our attention to different levels of blend complexity. To
deal with this issue in a more systematic way, we suggest the following categori-
sation regarding the input words used to form the name:
L1 each of the words appearing in the lexical blend is a commonly used word
for one input animal (no mapping);
L2 both input words represent input animal in a rather common way, but are
blended into one word by portmanteau principle, i.e. by using the prefix of
one word and the suffix of the other word (possibly with some intersection);
L3 one word represents one input animal with a commonly used word for this
animal, the other word represents a visible characteristic (part, colour etc.)
of the other animal (variant L3*: both words use such characteristics);
L4 one word represents one input animal with a commonly used word for this
animal, the other word represents a characteristic of the other animal for
which background knowledge about this animal (habitat, way of moving,
typical behaviour) is needed (variant L4*: both words use such characteris-
tics);
L5 one word represents one input animal with a commonly used word for this
animal, the other animal is represented with a more sophisticated association
– bisociation – for which a creative discourse into another realm (e.g. from
animals to literature) is needed (variant L5*: both words represented with
such associations).
We illustrate the categories by the names actually given in the survey to the
blended animal guinea bear :
L1 mouse-bear (input1: mouse, input 2: bear );
L2 rabbear (input1: rabbit, input 2: bear );
L3 small-headed bear (input1: mouse → small head, input 2: bear );
L4 scared bear (input1: mouse → scared, input 2: bear );
L5 mickey the bear (input1: mouse → Mickey the mouse, input 2: bear ).
As seen from this example, while the bear was easily recognised as one of
the constituting animals, there were different interpretations about the second
animal, “contributing” the head to the blended creature. In fact, the variety in
the whole dataset was even bigger as names given by different subjects suggested
the second animal being a mouse, rabbit, hamster, guinea pig, rat, squirrel,
wombat or opossum. The set of input words as used by the subjects is even
bigger since it includes also diminutives, slang versions, etc.
The levels increasing indicate the increasing complexity (but not necessarily
the quality) of the blends, but note that they do not build on just one criterion
in a linear way and there might also be a combination of principles described
at different levels present in one name. We illustrate this with a name teddybbit,
generated as a portamanteau (L2), but using an association between bear and
teddybear from the toys realm (L5).
98
However, we plan to improve this by introducing a creativity score in which
not just the level of mappings used will be taken into account, but also the fact
whether they were used for one or for both input animals, and how creative the
combination was (e.g., by taking into account phonetic features or by recognizing
references extrinsic to the two input animal spaces and their bisociations).
Note that not all of the names provided by the subjects in the survey were
lexical blends. Here we do not analyse such names in more detail, but to study
the potential for triggering creativity, they are important as well. Some examples
collected in our survey for the guinea bear are creepy, giant, or fluffy.
4 Patterns from examples for automated name generation
We investigated how the above-mentioned categories of human-generated names
could be used for automatic blend generation. Different categories represent dif-
ferent mechanisms. Names of Level 1 are very basic and easy to be automatically
generated, their creativity level is low and the name can hardy be called a blend.
On the other hand, higher levels (3-5) rely on human experience, background
knowledge, associations and bisociations. To generate the names of levels 3 and
4, we use a large web corpus (the enTenTen corpus [7]) and the sketch grammar
relations available in Sketch Engine [8]. For the last category (level 5), we used
other resources of human knowledge (Wikipedia, imdb lists). For each category,
we reveal the patterns in human given names and explain how they can be used
in automatic generation. Our generated examples are all done by modifying only
one animal name.
L1: In names given by humans, we found two different patterns at level
1. In each case, the two animals are used, the possible variations being either
hyphen to indicate the combined meaning “animal1-animal2” (e.g. dog-shark ) or
creating a single word containing full names of both animals “animal1animal2”
(e.g. spiderrat). The pattern with a premodifier of adjective can be recognised in
the given name mišasti medved, where the first word is an adjective formed from
the noun miš (Eng. mouse) and the second one is the noun medved (Eng. bear ).
Some word formations are language specific, e.g. in Slovene bare “noun-noun”
word formation is not very productive.
To illustrate the automatic name generation, we took the animal names from
each input space and concatenated them. Using these simple patterns resulted
in names very much resembling those generated by humans, e.g. duck-horse or
duckhorse. More examples are in the L1 row in Table 1.
L2: Level two uses the portmanteau principle. In all the languages used in
the survey this mechanism was used very frequently. For recognising these names
from the list, we focused on words composed of the beginning of one animal
word and ending of the other. Examples of basic portmanteau names given by
the subjects are the names given in Figure 1. We automatically recognized L2
blends by combining pairs of animals and some simple heuristics.
In automatic generation, the starting point was to combine half of the each of
the two input animal names. If the input word consists of two words, frequently
99
Table 1. Automatically generated names - examples for four fictional animals.
Input elephant & snake & horse & duck &
level chameleon horse chimpanzee horse
L1 elephant-chameleon snake-horse horse-chimpanzee, duck-horse
elephantchameleon snakehorse horsechimpanzee duckhorse
L2 elepheleon snarse horanzee ducrse
L3 tusk chameleon venom horse hoof chimpanzee beak horse
trunk chameleon fang horse mane chimpanzee arse horse
graveyard chameleon tail horse bridle chimpanzee back horse
tail chameleon poison horse rump chimpanzee feather horse
ear chameleon belly horse withers chimpanzee
L4 Asian chameleon venomous horse Trojan chimpanzee Anaheim horse
giraffe chameleon poisonous horse wild chimpanzee lame horse
captive chameleon garter horse Arabian chimpanzee Peking horse
L5 Dumbo chameleon Ser Hiss horse Alfonso chimpanzee Donald horse
Daffy horse Howard the horse
in the analysed examples one word is kept to from the blended name (which
is not a proper portmanteau anymore). This pattern was used for generating
examples like guinea lion, hammerhead eagle, hammerhead goose.
One could make different combinations based on different proportions of
the input words or by using phonetic rules (vowels, consonants, rhymes), exact
vs. inexact matching, pronunciation information, word’s Greek or Latin origins,
etc. as in many advanced existing systems proposing portmanteau name gener-
ation [19] [18] [3].
L3: In the next category of lexical blends, humans use visible characteristics
of one animal and associate them to the other animal. The properties of the an-
imal that gives the “head” to the new visual blend can be lexically expressed as
prepositional phrase modifying the head noun, i.e. the name of the animal pro-
viding the body (horse with snake head, elephant of the orange beak ), by adjec-
tive modifier (e.g. nosy robin, duckbilled pachyderm, trunkheaded chameleon)or
in noun-noun constructions (e.g. nosebird ). In some cases both animals are de-
scribed by their characteristic visible parts (e.g. tail-trunk ). Combinations with
portmanteau structure is also possible (e.g. grivasti kabod [Eng. mane horswan]).
For automated blend generation of L3 we currently use only noun-noun con-
structions. We rely on the Skecth Engine tool by using word sketches constructed
with Sketch grammar. Word sketches are automatic corpus-derived summaries
of a word’s grammatical and collocational behaviour [8]. From the word sketch
of animal “contributing” the head to the visual blend (e.g. elephant in Figure
1), we use all the collocators (above selected frequency and salience threshold)
100
from the grammatical category possessed. This lists contains nouns that in the
enTenTen corpus follow the search word and ’s, e.g. for elephant’s the list con-
tains tusk, trunk, ... resulting from collocation elephant’s tusk in the corpus.
We construct then noun-noun blends, by adding the animal name of the ani-
mal providing the body (e.g. chameleon). As shown in Table 1, examples using
this structure often correspond to parts of the body, (tusk chameleon, trunk
chameleon, tail chameleon, ear chameleon), while graveyard chameleon does not
represent the part of the body. Obviously, some of the compounds are irrelevant,
e.g. tail chameleon – since chameleons have a tail themselves so this description
does not contribute anything in terms of blending. Neither does the corpus pro-
vide the information if the “possessed” part is located on the animal’s head and
even less if it corresponds to the depicted picture (e.g. tusks are not depicted on
the picture of elephant and chameleon from Fig. 1, even if they are prototypical
part of elephant’s head). More specific filters and knowledge bases will be used
in future to narrow the choice to better candidates.
L4: Level 4 names are more diverse and require more background knowl-
edge. As mentioned in Section 3, the observed categories are habitat, locomotion
(plavajoči konj [Eng. swimming horse], typical behaviour (e.g.elequack using an-
imal sounds) or usage (saddleducks. Again, also both animals can be represented
by their properties, such as in the blended name galloping quack. For automated
name generation at this level, we used again the word sketches, but we took
the information from category modifiers (typical adjectival or noun collocators
modifying the animal providing the head to the blended creature). E.g. adjec-
tives venomous and poisonous are typical collocators of word snake and are used
for forming blended names venemous horse and poisonous horse. Often breed
names are used in modifier position; by selecting only lower case modifiers we
can keep more general properties. For Level 4 , more background knowledge
is needed. E.g., from automatically constructed names Trojan chimpanzee, wild
chimpanzee or Arabian chimpanzee, the first one is referring to specific cultural
reference Trojan horse and can be interpreted at level 5. Same goes for the lame
horse, which is formed from the idiom lame duck (i.e.an elected official who
is approaching the end of his tenure, and esp. an official whose successor has
already been elected (Wikipedia)).
L5: In analysis of human lexical blends we manually classified in Level 5 the
bisociative blends using characters from cartoons (Spider Gonzalez ), children
songs (Slonček Raconček refering to a Slovene song Slonček Jaconček ), where
slonček means small elephant and raconcek comes from duck – raca), movies (
My little mallard ), politicians (Sharkozy), legends (Jezerski Pegasus [Eng. river
Pegasus]) and often combinations of several of them, e.g. character from movie
and from comic strips Jumbo Zvitorepec (where Jumbo refers to the animal,
while Zvitorepec is a character from Slovene comic strip by Miki Muster, but
literally means curled tail which refers also to the visual representation of this
animal (cf. picture elephant, chameleon in Fig. 1).
For automatically generating highly creative lexical blends inspired by the
examples given by participants, we based the bisociative blend generation on
101
characters from the movies representing the input animal. We created a short
list from Wikis, IMDB and Wikipedia pages about animal characters in movies
where the last section covers cultural representations. In the name generation
process, we first checked if character’s name contains the name of the animal and
if so we substituted this name with the name of the other input animal (e.g. horse
substituting the duck in Donald horse). On the other side, if the animal does
not appear explicitly we added the name of the second animal to the existing
character name (Dumbo chameleon). In future, we will expand generation of
names at this level by exploring other realms besides movies and books.
5 Discussion
We investigated the principles of creating lexical blends based on visual blends
(blended animals). We revealed different mechanisms used in name formation
and introduced a new categorisation of blend complexity (L1-concatenation
blends; L2-portmanteaux; L3-blending based on visible characteristics; L4- blend-
ing using background knowledge and L5-bisociative blends). After the analysis
of examples generated names by humans, we made a prototype system for au-
tomated generation of blends of different levels using word combinations, gram-
matical and collocational information and background knowledge resources. The
most frequent mechanism used by humans was the portmanteau principle. But
a portmanteau can vary from very basic ones to the bisociative ones, since blend
strategies can easily be combined. For instance, the blend shagull can be in-
terpreted as a simple portmanteau blend (shark+gull) or as bisociative blend
refering to Chagall. This example shows that the bisociation can be used on the
production level (e.g. creative blend but the reader cannot decompose it), on the
interpretation level (e.g. even if there was no such intention when generating a
name, the bisociation can be present at the reader’s side) or both.
We like some names generated as lexical blends more than the others – what
counts? Even if names are generated using similar principles, some of them are
much more creative, achieving higher degree of creative duality, compressing
multiple levels of meaning and perspective into a simple name [20]). It is the
combination of simplicity and bisociation (in our case the switch from animal
wor(l)d to cultural realm) that seems to be the most impressive. To verify this
claim and to get a more thorough evaluation of automatically generated names,
we plan to collect human subjects feedback as well as compare human-generated
and automatically generated names. We will also further elaborate the automatic
recognition of blend complexity and on the other side the blend generation part
(e.g. including phonological criteria, rhymes, more background knowledge, etc.).
Next, we will investigate the role of emotions: while some names were neutral,
many had very strong emotional content (cf. negative emotions in disgusoarse,
horrabit or the name given to the hammerhead gull, for which instead of naming
it a user wrote “deserves death by fire, not a name”) or positive emotions in le
trop joli, name used for guinea lion. Another spectre of research is to investigate
the generality of our blend categorisation by applying it to other domains.
102
This work has been supported by the projects ConCreTe (grant nb. 611733), WHIM
(611560) and Prosecco (600653) funded by the European Commission, FP 7, the ICT
theme, and the FET program. We thank also A. Fredriksen, the author of the pictures.
References
1. Bhatta, S.R., Goel, A.: Learning generic mechanisms for innovative strategies in
adaptive design. The Journal of the Learning Sciences 6, 367–396 (1997)
2. Boden, M.: The Creative Mind: Myths and Mechanisms. Basic Books (1991)
3. Deri, A., Knight, K.: How to make a frenemy: Multitape FSTs for portmanteau
generation. In: Proc. of the NAACL (2015)
4. Domeshek, E., Kolodner, J.: A case-based design aid for architecture. In: Gero, J.,
Sudweeks, F. (eds.) Artificial Intelligence in Design ’92, pp. 497–516 (1992)
5. Fauconnier, G., Turner, M.: Conceptual integration networks. Cognitive Science
22(2), 133–187 (1998)
6. Goel, A.K., Craw, S.: Design, innovation and case-based reasoning. The Knowledge
Engineering Review 20(3), 271–276 (2005)
7. Jakubı́ček, M., Kilgarriff, A., Kovář, V., Rychlỳ, P., Suchomel, V., et al.: The
tenten corpus family. In: 7th Int. Corpus Linguistics Conf. pp. 125–127 (2013)
8. Kilgarriff, A., Rychly, P., Smrz, P., Tugwell, D.: The sketch engine. In: Proc. EU-
RALEX 2004. pp. 105–116 (2004)
9. Koestler, A.: The Act of Creation. New York:Macmillan (1964)
10. Kolodner, J.: Case-Based Reasoning. Morgan Kaufmann Publishers Inc. (1993)
11. Kuznetsova, P., Chen, J., Choi, Y.: Understanding and quantifying creativity
in lexical composition. In: Proc. Conf. Empirical Methods in Natural Language
(EMNLP). pp. 1246–1258 (2013)
12. Lehrer, A.: Blendalicious. Lexical creativity, texts and contexts pp. 115–133 (2007)
13. Li, B., Zook, A., Davis, N., Riedl, M.O.: Goal-driven conceptual blending: A com-
putational approach for creativity. In: Proc. of the 3rd Int. Conf. on Computational
Creativity. pp. 9–16 (2012)
14. Martins, P., Urbančič, T., Pollak, S., Lavrač, N., Cardoso, A.: The good, the bad,
and the AHA! blends. In: Proc. of the 6th Int. Conf. on Computational Creativity
(2015)
15. Norman, D.A.: The design of everyday things. Basic books, NY (1988)
16. Özbal, G., Strapparava, C.: A computational approach to the automation of cre-
ative naming. In: Proc. of the 50th Annual Meeting of the Association for Com-
putational Linguistics: Long Papers-Volume 1. pp. 703–711 (2012)
17. Pereira, F., Cardoso, A.: The horse-bird creature generation experiment. The
Interdisciplinary Journal of Artificial Intelligence and the Simulation of Be-
haviour(AISBJ ) 1(3), 257–280 (2003)
18. Smith, M.R., Hintze, R.S., Ventura, D.: Nehovah: A neologism creator nomen ip-
sum. In: Proc. of the 5th Int. Conf. on Computational Creativity. pp. 173–181
(2014)
19. Veale, T.: Tracking the lexical zeitgeist with WordNet and Wikipedia. In: Proc.
European Conf. on Artificial Intelligence (ECAI). pp. 56–60 (2006)
20. Veale, T., Butnariu, C.: Harvesting and understanding on-line neologisms. Cogni-
tive Perspectives on Word Formation pp. 399–420 (2010)
21. Wiggins, G.A.: Towards a more precise characterization of creativity in AI. In:
Proc. of Workshop Program of ICCBR – Creative Systems: Approaches to Cre-
ativity in AI and Cognitive Science (2001)