"Talent, Skill and Support." A Method for Automatic Creation of Slogans

Slogans are an effective way to convey a marketing message. In this paper, we present a method for automatically creating slogans, aimed to facilitate a human slogan designer in her creative process. By taking a target concept (e.g. a computer) and an adjectival property (e.g. creative) as input, the proposed method produces a list of diverse expressions optimizing multiple objectives such as semantic relatedness, language correctness, and usage of rhetorical devices. A key component in the process is a novel method for generating nominal metaphors based on a metaphor interpretation model. Using the generated metaphors, the method builds semantic spaces related to the objectives. It extracts skeletons from existing slogans, and finally fills them in, traversing the semantic spaces, using the genetic algorithm to reach interesting solutions (e.g. “Talent, Skill and Support.”). We evaluate both the metaphor generation method and the overall slogan creation method by running two crowdsourced questionnaires.

[1]  Gözde Özbal,et al.  BRAINSUP: Brainstorming Support for Creative Sentence Generation , 2013, ACL.

[2]  G. Steen,et al.  Making ads less complex, yet more creative and persuasive: the effects of conventional metaphors and irony in print advertising , 2015 .

[3]  Hairong Li,et al.  What Makes a Slogan Memorable and Who Remembers It , 1994 .

[4]  S. Glucksberg Understanding figurative language : from metaphors to idioms , 2001 .

[5]  Sarah Harmon,et al.  FIGURE8: A Novel System for Generating and Evaluating Figurative Language , 2015, ICCC.

[6]  Gonzalo Méndez,et al.  Exploring the Role of Word Associations in the Construction of Rhetorical Figures , 2016, ICCC.

[7]  Josep Blat,et al.  Generating Apt Metaphor Ideas for Pictorial Advertisements , 2013, ICCC.

[8]  Tony Veale,et al.  Creating Similarity: Lateral Thinking for Vertical Similarity Judgments , 2013, ACL.

[9]  Stefan Evert,et al.  Corpora and collocations , 2007 .

[10]  Boris Kabanoff,et al.  Computer-aided content analysis: What do 240 advertising slogans have in common? , 1996 .

[11]  Walter Daelemans,et al.  Pattern for Python , 2012, J. Mach. Learn. Res..

[12]  Khalid Al-Najjar,et al.  Meta4meaning: Automatic Metaphor Interpretation Using Corpus-Derived Word Associations , 2016, ICCC.

[13]  R. Giora On Our Mind: Salience, Context, and Figurative Language , 2003 .

[14]  Martin Znidarsic,et al.  Implementation of a Slogan Generator , 2014, ICCC.

[15]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[16]  G. Tom The Use of Rhetorical Devices in Advertising , 1999 .

[17]  Hannu Toivonen,et al.  Expanding and Weighting Stereotypical Properties of Human Characters for Linguistic Creativity , 2017, ICCC.

[18]  Marc Parizeau,et al.  DEAP: evolutionary algorithms made easy , 2012, J. Mach. Learn. Res..

[19]  Darryl W. Miller,et al.  An analysis of rhetorical figures and other linguistic devices in corporation brand slogans , 2016 .

[20]  I. Richards The Philosophy of Rhetoric , 1936 .

[21]  Matthew Purver,et al.  From Distributional Semantics to Conceptual Spaces: A Novel Computational Method for Concept Creation , 2015, J. Artif. Gen. Intell..

[22]  Carlo Strapparava,et al.  Automatizing Two Creative Functions for Advertising , 2007 .

[23]  Yair Neuman,et al.  Literal and Metaphorical Sense Identification through Concrete and Abstract Context , 2011, EMNLP.