Automatic Recognition of Personality from Digital Annotations

Nizar Omheni

, Anis Kalboussi

, Omar Mazhoud

and Ahmed Hadj Kacem

Higher Institute of Computer and Management, University of Kairouan, Kairouan, Tunisia

Faculty of Economics and Management, University of Sfax, Road of the Airport Km4 3018, Sfax, Tunisia

Keywords: Learning User Profile, Data Extraction, Annotations, Digital Annotations, Personality Traits.

Abstract: There is an increasing interest in understanding human perception based on reading and writing behaviours.

Such researches are interested to seek knowledge of an individual’s personality as a way to predict their

behaviours and preferences across different contexts and environments. Recent works show significant

relation between the reader personality and his reading behaviours. Based on these findings, annotation

activity is considered as potential source to predict certain personality traits of readers. In this paper, we take

advantage of such theoretical works and we propose an online environment of active reading used to

explore practically the utility of annotation in reflecting an accurate user personality profile. We apply the

paired t-test to evaluate the system’s efficiency to measure human traits versus the scores of personality

traits measured using the Neo-IPIP inventory. Our findings show plainly that some traits of users’

personalities can be predicted accurately from digital annotation traces during online reading session.

1 INTRODUCTION

The psychological researches show that personality

traits are consistently stable over time and constitute

a significant inter-individual difference (Burger,

2011; Cobb-Clark and Schurer, 2012). This stability

is considered as the key-assumption of personality

psychology which has the aim to predict observable

individual differences based on human traits

characteristics that is measurable in quantitative

terms (Matthews et al., 2009). The findings of

personality psychology area interested increasingly

the computing community which leads to the growth

in number of research paper in the topic of

personality computing (Vinciarelli and Mohammadi,

2014).

Many scholars are motivated to model users’

traits and they are interested to seek knowledge of an

individual’s personality as a way to predict their

behaviours and preferences across different contexts

and environments (Bologna et al., 2013; Selfhout et

al., 2010).

To assess user’s personality traits, several works

control user’s behavioural residues traces in the

digital environments (Kosinski et al., 2013). For

instance, (Bachrach et al., 2012; Golbeck et al.,

2011; Quercia et al., 2011) have analyzed the

relationship between personality and users’

behaviours in on-line social environments. The

works’ findings show that users’ personalities can be

accurately predicted through their traces in social

profiles.

There is an increasing interest in understanding

human perception based on features extracted from

reading and writing behaviours. To show the ability

to profile user personality from human text

production and peculiarities of reading behaviours,

many researchers study the relation between user’s

personality traits and several factors such as text

(Wright and Chin, 2014; Celli, 2012; Arya et al.,

2012), linguistic cues (Celli and Poesio, 2014;

Mairesse et al., 2007; Celli, 2012a), handwriting

styles (Rahiman et al., 2013; Prasad et al., 2010;

Fisher et al., 2012), posts written in online social

spaces (Iacobelli et al., 2011; Sumner et al., 2012;

Qiu et al., 2012), social annotation (Mezghani et al.,

2012) and annotation traces during active reading

sessions (Omheni et al., 2014; Jackson, 2001).

The present work focuses on development of an

automated technique for determining the personality

traits of a user through analysis of digital annotation

traces in online reading environment.

The rest of this paper is as follows. In the next

section, we present an overview on related works.

Then, we propose an automated method to predict

accurately certain traits of users through their

annotation practices. Thereafter, we evaluate the

system’s performance to measure precisely the

273

Omheni N., Kalboussi A., Mazhoud O. and Kacem A..

Automatic Recognition of Personality from Digital Annotations.

DOI: 10.5220/0005483002730280

In Proceedings of the 11th International Conference on Web Information Systems and Technologies (WEBIST-2015), pages 273-280

ISBN: 978-989-758-106-9

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

scores of users’ traits. Finally, we draw some

conclusions and we cite certain possible directions

for future work.

2 RELATED WORKS

The computing community is interested increasingly

to seek knowledge of an individual’s personality as a

way to predict their behaviours and preferences

across different contexts and environments

(Vinciarelli and Mohammadi, 2014).

The human personality is viewed from different

perspectives: biologically, psychoanalytically,

behaviourally, humanistically, cognitively, trait

perspective, etc. The most accepted model in

computing area is the trait perspective. Indeed, the

trait based model represents personality in terms of

numerical values which is a form particularly

suitable for computer processing (Vinciarelli and

Mohammadi, 2014).

Commonly, personality traits are assessed using

self-report techniques (Boyle and Helmes, 2009). In

computing area, the most popular technique used to

assess personality traits is the NEO-Personality-

Inventory, where the user rates his own behaviour

with Likert scales. For instance, (Nunes et al., 2008;

Nunes et al., 2008a) propose obtaining users’

personality information through their answers to the

NEO-IPIP inventory. The participators in the

authors’ experimentation were instructed to answer

900 questions. 10% of people answered all the

questions of the testing because they aren’t ready to

spend much effort for a long time to complete a

multi-item questionnaire. (Hu and Pu, 2010) use the

TIPI test (Ten-Item Personality Inventory)

developed by (Gosling et al., 2003) to acquire the

user’s personality characteristics. This inventory is

an extremely brief measure of the Big-Five

personality dimensions so the acquisition process

takes about 2-3 minutes to complete. Further

research works follow the same path to calculate the

user personality scores where each subject was

instructed to fill in a big five questionnaire (Tkalcic

et al., 2009; Wu et al., 2013).

Although the results shown in previous works is

fruitful we believe these researches have left certain

open issues concerning the followed approach to

obtain the information needed in the user modelling

process. Indeed, the crucial constraint in the

profiling process is to model a credible user’s profile

which reflects truly as much as possible the user in

the working environment. The explicit methods

require much from the user who is not ready,

usually, to fill long forms or even to write the truth

when completing forms about themselves

(Schiaffino and Amandi, 2009). Therefore, the main

limitation of self-assessments technique is that the

users might tend to bias the ratings towards socially

desirable characteristics knowing that the web-based

psychometric tests suffer of the control diminution

over the testing situation which lead to the high

probability of cheat especially in case where the

motivation to do is obvious and the personality

assessment can have negative consequences like,

e.g., failing a job inter

view (Vinciarelli and

Mohammadi, 2014; Barak et al., 2004; Gawronski

and De Houwer, 2014).

Certain psychologists seek to alternative

measurement instruments that reduce participants’

ability to control their responses and do not require

introspection for the assessment of psychological

attributes (Gawronski and De Houwer, 2014).

According to Brunswick’s lens model human

personality is externalized through distal cues

observable by others. These distal cues are

essentially physical traces left by individuals’

behaviours in virtually everything observable they

do (Vinciarelli and Mohammadi, 2014). In this

context, recent studies show the opportunity to

derive personality from digital traces of human

behaviours in different workspaces. Such works are

interested to show how users’ behaviours on Social

Networks relates to their personality, as measured by

the standard Five Factor Model (Kosinski et al.,

2013; Bachrach et al., 2012; Golbeck et al., 2011).

Other scientists study recognizing personality in

user’s speech and social interaction (Polzehl et al.,

2011; Ivanov et al., 2011).

Actually, in personality computing area, there is

a great interest to understand the human perception

based on reading and writing behaviours (Rahiman

et al., 2013; Wright and Chin, 2014; Minamikawa

and Yokoyama, 2011).

In the current essay, we are interested to a

specific behaviour of reading and writing activity:

The annotations. Indeed, we conducted a previous

work to show the relation between reader’s

personality and his annotation practices in “pen-and-

paper” context. The experiment showed an

interesting relation of correlation between certain

peculiarities of annotation activity and the

annotator’s personality traits (Omheni et al., 2014).

These findings motivate us to consider

the annotation traces in the digital context and to test

the possibility to recognize human’s personality

traits based on their digital annotations’ traces in

online reading environment.

WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies

274

3 ANNOTATION AND

PERSONALITY

Annotation is a handwritten practice which bridges

between reading and writing (Marshall, 1998) and

constitutes the most prominent habits of active

reading activity (Thomas, 2007).

(Kirwan, 2010, p. 5) considers the reader

marginalia (annotation) as: the “most direct,

reactionary response to the text that can feasibly be

considered” to study the relation between the reader

identity and the text. According to (Kirwan, 2010)

the annotations provide the link between reader, text,

and meaning and reflect the subjective individuality

of the annotator’s responses to the text. Based on

this subjective relationship, the author suggests

expanding the psychology-based reader theory to

include reader’s annotation practices.

The annotation activity is “a basic and often

unselfconscious way in which readers interacts with

texts” (Marshall, 2009, p. 38). Furthermore, the

annotation is described as a natural human activity

that is used in daily life as an integral part of reading

activity (O’hara and Sellen, 1997). Every annotator

has unique individual patterns in making annotations

(Naghsh, 2007). According to (Jackson, 2001, p. 5),

“if you ask annotators today what systems they use

for marking their books and where they learned

them, they generally tell you that their methods are

private and idiosyncratic”. Hence, the individuality

of annotation patterns shows us very plainly that

there can be some sort of connection between

annotation practices and annotator’s personality.

(Jackson, 2001) assumes that “marginalia

[annotation] express a reader’s impulsive and

unguarded reactions to a book” and she “consider[s]

them to be an exceptionally reliable guide to

personality” (Jackson, 2001, p. 87).

In our prior work (Omheni et al., 2014) we

conducted an empirical study to show the implicit

relation between reader’s annotation activity and his

personality traits. The study shows significant

correlations for Neuroticism, Conscientiousness, and

Extraversion traits. Furthermore, we made

predictions about a subject’s personality based on

multiple annotation features using the multivariate

linear regression method. Our findings show that

Neuroticism and Conscientiousness can be predicted

with reasonable accuracy, whereas other traits are

more difficult to be predicted.

Regarding the external validity of our findings,

first of all, in our study, we addressed the population

characterized as following: people aged between 18

and more, either man or woman, with different

occupations and interests and who practice

frequently the habit of reading and annotation of

textual materials. Hence, the size of population is so

large, which is in reality, not possible to sample the

whole population, due to budget, time and

feasibility. Thus, as a way to decide pragmatically

the generalization of our findings we made on the

basis of the selection of a sample group that is

representative of the target population. This is

something that we took into account when designing

our experiments. Our research design is governed by

the interest in the generalisability of our study’s

results. Hence, we do a good job of drawing a

representative sample from our addressed population

and we have not considered specific circumstances

of time and place in which the data were collected.

Indeed, in our study, we focus on the annotation

practice which is a ubiquitous human behaviour and

we have only considered the paper support of

annotated document. To measure the personalities

scores of volunteers we utilized the standard Five

Factor Model questionnaire (the NEO-IPIP

Inventory) which is the best accepted and most

commonly used scientific measure of human

personality traits (Peabody and De Raad, 2002). All

the tools and circumstances taken into account to

achieve our experiment can be considered in a

variety of places, with different people and at

different times. Thus, based on logical

considerations and speculation concerning the extent

to which our sample is similar to the target

population and the replication of our experiments

with other representative groups in other locations

can strongly give the same results and findings, we

ensure that we can generalize our findings to the

entire population in our study.

4 RECOGNITION OF READER

TRAITS BASED ON DIGITAL

ANNOTATION PRACTICES

Based on what previously cited, it is plain, that a

reader’s annotation activity is really an expression of

his personality traits. Indeed, we show very plainly

that the considered features descriptive of the

annotation practices in our prior study may appear

insignificant in themselves, but, they are

nevertheless all very significant as indications of the

annotator’s personality traits (Omheni et al., 2014).

Recent researches endeavour to replace the “pen-

and-paper” paradigm for the annotation needs by

employing the technology of free form digital ink

AutomaticRecognitionofPersonalityfromDigitalAnnotations

275

annotations which add the flexibility and natural

expressiveness of the traditional handwriting method

to the digital annotation process (Kalboussi et al.,

2015). Such tools enable readers to annotate their

digital documents similarly to “pen-and-paper” case.

For instance, iAnnotate (Plimmer et al., 2010) is an

annotation tool for android system which enables

users to add annotations with the pencil, highlighter,

and note tools to their digital texts. Hence, the digital

context of free form annotation process is very

similar to the context of pen-and paper. The high

degree of proximal similarity among these two

contexts constitutes a strong evidence to generalize

our study’s results (Omheni et al., 2014) to digital

annotation environment. Thus, we are motivated to

take advantage of digital annotations which can be

considered as a source of knowledge to

automatically predict an annotator’s personality

traits.

In this subpart of our study, we focus on

development of an automated technique for

determining the characteristic traits of a person

through the features extracted from his annotation

activity. In fact, the proposed system “i-Read” is an

online reading environment where the user can

upload their reading materials and practice their

annotation habit.

The following figure (fig.1) illustrates the

interaction between the various modules of “i-Read”

system along with the flow of information/data. The

proposed architecture consists of user annotation

interface, the annotation analyzer module, the

constructor profile module and three databases with

two servers.

Figure 1: The Architecture of “i-Read” Online Reading

Environment.

To avoid destroying the original version of

reading materials, our system uses an independent

annotation database, which differs from the

documents database, to store annotated information

and contexts from readers. Moreover, the annotation

interface provides several powerful annotation

functionalities, such as scribbling, highlighting,

underlining, commenting, as a way to engage users

actively with their reading materials.

4.1 The Annotation Analyser Module

To predict the actual personality traits of the

individual we consider the features studied in

(Omheni et al., 2014) to be extracted automatically

by the module of annotation analysis. To extract the

considered features we start, first of all, by

classifying annotations in three general categories.

This categorization is based on how annotations can

appear and be represented. (Agosti and Ferro, 2003)

define three ways to represent the meaning of

annotation:

1. Textual annotation expressed by a piece of

text added to the annotated document,

2. Graphic annotation expressed by a graphic

mark added to a document,

3. Reference annotation expressed by a link

between two texts or two textual pieces in the

same document.

The authors called these basic ways “signs of

annotation” and they define the term sign as a

formation of a meaning. Furthermore, according to

(Agosti and Ferro, 2003), these signs can be

combined together to express more complex signs of

annotation. In current work we try to quantify the

user’s digital annotation traces by collecting a set of

statistics describing the total number of annotations,

average number of annotations per page of reading

material, number of textual annotations, number of

graphical annotations, and number of referential

annotations.

4.2 The Constructor Profile Module

In our empirical study, we used the multiple linear

regression analysis to assess the association between

six independent variables representing the different

features qualifying the annotation activity and a

single continuous dependent variable represents the

focused user’s trait. The multiple linear regression

equation is as follows:

Y = b

+ b

(1)

WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies

276

where Y is the predicted or expected value of the

dependent variable representing the score of the

focused user’s personality trait, X

through X

are the

distinct independent or predictor variables, b

is the

value of Y when all of the independent variables (X

through X

) are equal to zero, and b

through b

are

the estimated regression coefficients. Based on this

mean function, we can determine the expected

annotator’s personality trait as long as we know

certain peculiarities characterizing quantitatively her

annotation practices.

As we cited previously, based on the similarity

between the two contexts of annotation process (the

manual and digital contexts) we have generalized

our study’s results to the digital environment. Thus,

in current work, we are motivated to apply the

study’s findings (Omheni et al., 2014). The scenario

of the proposed automatic personality profiling is as

follow: first of all the annotation analyzer module

captures certain features which tend to characterize

quantitatively the reader’s annotation practices. The

collected data is used thereafter as a source of

knowledge to extract the implicit information which

describes the personality of active user. Indeed, the

quantitatively information is transferred to the

constructor profile module as an input data to the

equations used to estimate the scores of user traits

profile.

4.3 System Operation Procedure

Based on the system architecture (fig.1), the system

operating procedure is described and summarized as

follows.

1. A user uploads his/her reading document on

the “i-Read” online environment;

2. The system saves the document in the

documents repository;

3. The user annotates his/her reading material;

4. The system saves the user’s annotations in

the Annotations repository;

5. The annotation analyzer module captures the

user’s annotation activity and extracts such

features;

6. The annotation analyzer module sends the

collected information to the profile

constructor module to build the user

personality traits profile;

7. The profile constructor module considers the

collected information as an input data to the

equations used to estimate the scores of user

traits profile;

8. The system saves the modelled user’s profile

in the Profiles repository.

4.4 System’s Performance Evaluation

In order to evaluate the system’s level of

performance in measuring accurately the scores of

reader’s personality traits, we conducted the

following experiment and we invited 100 volunteers.

The invited people have the same characteristics

qualifying the target population in our prior work

(age, gender, habit of reading and annotation).

To assess whether our system measures accurately

the user’s traits, we invited the participators to

upload their textual materials on the “i-Read”

environment and we instructed them to use the

system to achieve their reading and annotation

activities (fig.2). Next, they were instructed to

answer a standard Five Factor Model questionnaire

(the NEO-IPIP Inventory) to obtain a feedback

regarding their personality based on their responses.

Figure 2: Annotated document opened in “i-Read” online

environment.

To show the system’s efficiency to measure

accurately the scores of reader’s conscientiousness

and neuroticism traits compared to the values

determined using the NEO-IPIP Inventory, we

applied the paired t-test to compare the scores of

certain user’s personality traits obtained through the

two different methods of measurement. We look to

determine whether there is a significant difference

between the paired values of scores. Both

measurements are made on each subject in the

selected sample, and the test is based on the paired

differences between these two values. The test

statistic is calculated as following:

t =

/√s

/n (2)

 is the mean difference, s

is the sample variance, n

is the sample size and t is a Student t quantile with

n-1 degrees of freedom. In our case n = 100. Tables

3 and 4 show descriptive statistics of t-test measure

of the difference significance between the paired

AutomaticRecognitionofPersonalityfromDigitalAnnotations

277

values of user’s conscientiousness and neuroticism

traits scores measured with two different systems:

the “i-Read” system and the Neo-IPIP inventory.

Analytical results indicate that the scores of user’s

Conscientiousness and Neuroticism characteristics

obtained through the “i-Read” system did not differ

significantly (Sig1 = 0,72 > 0.05; Sig2 = 0,53 >

0.05) versus the scores measured using the Neo-IPIP

inventory (Table 1. and 2.). Thus, the experimental

results show the possibility to measure some

personality traits (Conscientiousness and

Neuroticism) with reasonable accuracy by reference

to reader’s digital annotation practices.

Table 1: A t-test measure of the difference significance

between the paired values of Conscientiousness scores

measured with two different systems.

Scores

measured

with

Mean Std.Dv. t-value p-value

“i-Read”

system

25,78 4,90

Neo-IPIP

inventory

26,50 20,25 -0,36 0,72

Table 2: A t-test measure of the difference significance

between the paired values of Neuroticism scores measured

with two different systems.

Scores

measured

with

Mean Std.Dv. t-value p-value

“i-Read”

system

64,66 6,74

Neo-IPIP

inventory

63,37 21,16 0,63 0,53

These results is coherent to our prior findings

(Omheni et al., 2014) and support the hypothesis of

the existence of some sort of connection between

annotation traces and certain personality traits of the

annotator. These discoveries are promising and

constitute a new tendency in modelling human traits

by reference to certain behavioural residues of

reading and writing activities.

5 CONCLUSIONS

In this paper, an automated method has been

developed to predict certain personality

characteristics of a person from features extracted

through his annotation behaviours. We have proven

that some traits of users’ personalities can be

predicted accurately from digital annotation traces in

online reading environment. This shows us very

convincingly that there is some sort of connection

between the peculiarities of annotation activity and

certain personality traits of the annotator.

We want to enhance the proposed system to be

used in personality-based systems. Another future

direction, we want to increase the features used to

derive user’s personality traits based on their

annotation practices for more correct results. These

investigations can be subjects of follow-up works in

the near future.

The current work can be complementary to the

prior works of (Kalboussi et al., 2014; Kalboussi et

al., 2013; Kalboussi et al., 2013a; Kalboussi et al.,

2013b) which aim to invoke the web services based

on annotation activity of a reader. Thus, our work

can be useful in the adaptation process of the

invoked web services. As a summary, we can

consider the modelling process of reader's

personality based on annotation traces is a step

forward in the area of personalization over web.

REFERENCES

Agosti, M. and Ferro, N. (2003). Annotations: Enriching a

digital library. In Research and Advanced Technology

for Digital Libraries, pages 88–100. Springer.

Arya, A., Ragini, S., Kumar, H., Abinaya, G., et al.

(2012). A text analysis based seamless framework for

predicting human personality traits from social

networking sites. International Journal of Information

Technology and Computer Science (IJITCS), 4(10):29.

Bachrach, Y., Kosinski, M., Graepel, T., Kohli, P., and

Stillwell, D.(2012). Personality and patterns of

facebook usage. In proceedings of the 3rd annual

ACM web science conference, pages 24–32. ACM.

Barak, A., Buchanan, T., Kraus, R., Zack, J., and Stricker,

G. (2004). Internet-based psychological testing and

assessment. Online counseling: A handbook for mental

health professionals, pages 217–239.

Bologna, C., De Rosa, A. C., De Vivo, A., Gaeta, M.,

Sansonetti, G., and Viserta, V. (2013). Personality

based recommendation in e-commerce. In UMAP

Workshops. Citeseer.

Boyle, G. J. and Helmes, E. (2009). Methods of

personality assessment.

Burger, J. (2011). Personality. Editor Wadsworth, USA,

edition.

Celli, F. (2012). Adaptive Personality Recogntion from

Text. PhD thesis, University of Trento.

Celli, F. (2012a). Unsupervised personality recognition for

social network sites. In ICDS 2012, The Sixth

International Conference on Digital Society, pages

59–62.

Celli, F. and Poesio, M. (2014). Pr2: A language

independent unsupervised tool for personality

WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies

278

recognition from text. arXiv preprint

arXiv:1402.2796.

Cobb-Clark, D. A. and Schurer, S. (2012). The stability of

big-five personality traits. Economics Letters,

115(1):11–15.

Fisher, J., Maredia, A., Nixon, A., Williams, N., and Leet,

J. (2012). Identifying personality traits, and especially

traits resulting in violent behavior through automatic

handwriting analysis.

Gawronski, B. and De Houwer, J. (2014). Implicit

measures in social and personality psychology.

Handbook of research methods in social and

personality psychology, 2.

Golbeck, J., Robles, C., Edmondson, M., and Turner, K.

(2011). Predicting personality from twitter. In Privacy,

security, risk and trust (passat), 2011 ieee third

international conference on and 2011 ieee third

international conference on social computing

(socialcom), pages 149–156. IEEE.

Gosling, S. D., Rentfrow, P. J., and Swann, W. B. (2003).

A very brief measure of the big-five personality

domains. Journal of Research in personality,

37(6):504–528.

Hu, R. and Pu, P. (2010). Using personality information in

collaborative filtering for new users. Recommender

Systems and the Social Web, page 17.

Iacobelli, F., Gill, A. J., Nowson, S., and Oberlander, J.

(2011). Large scale personality classification of

bloggers. In Affective Computing and Intelligent

Interaction, pages 568–577. Springer.

Ivanov, A. V., Riccardi, G., Sporka, A. J., and Franc, J.

(2011). Recognition of personality traits from human

spoken conversations. In INTERSPEECH, pages

1549–1552.

Jackson, H. (2001). Marginalia: Readers Writing in

Books. Nota Bene series. Yale University Press.

Kalboussi, A., Omheni, N., Mazhoud, O., Kacem, A.H.

(2015). ‘How to Organize the Annotation Systems in

Human-Computer Environment: Study, Classification

and Observations’,In: Proceedings of the 15

IFIP

TC.13 International Conference on Human-Computer

Interaction - INTERACT 2015, Bamberg, Germany,

Springer.

Kalboussi, A., Mazhoud, O., Omheni, N., Kacem, A.H.

(2014). ‘A New Annotation System based on a

Semantic Analysis of Learner’s Annotative Activity to

Invoke Web Services’,International Journal of

Metadata, Semantics and Ontologies (IJMSO), Vol.9

(4),Inderscience, pp.350–370.

Kalboussi, A., Mazhoud, O., Omheni, N., Kacem, A.H.

(2013).‘An Approach of Assistance of Learner’s

Annotative Activity through Web Services’,

Australian Journal of Intelligent Information

Processing Systems (AJIIPS),Vol.13 (3), Special Issue

on Edutainment’13, pp.15–22.

Kalboussi, A., Mazhoud, O., Hadj Kacem, A., Omheni, N.

(2013a).‘A Formal Model of Learner’s Annotations

Dedicated to Web Services Invocation’, In:

Proceedings of the 21

International Conference on

Computers in Education (ICCE’13), APSCE,pp.166–

169.

Kalboussi, A., Mazhoud, O., Kacem, A.H. (2013b).

‘Annotative Activity as a Potential Source of Web

Service Invocation’,In: Proceedings of the 9

International Conference on Web Information Systems

and Technologies (WEBIST’13), SciTePress, pp.288–

292.

Kirwan, K. (2010). Rebellious readers: A marginalized

reader theory of identity, text, and meaning.

Kosinski, M., Stillwell, D., and Graepel, T. (2013). Private

traits and attributes are predictable from digital records

of human behavior. Proceedings of the National

Academy of Sciences, 110(15):5802–5805.

Marshall, C. C. (1998). Toward an ecology of hypertext

annotation. In Proceedings of the ninth ACM

conference on Hypertext and hypermedia: links,

objects, time and space—structure in hypermedia

systems: links, objects, time and space—structure in

hypermedia systems, pages 40–49. ACM.

Marshall, C. C. (2009). Reading and writing the electronic

book. Synthesis lectures on information concepts,

retrieval, and services, 1(1):1–185.

Matthews, G., Deary, I. J., and Whiteman, M. C. (2009).

Personality Traits. Cambridge University Press.

Mezghani, M., Zayani, C. A., Amous, I., and Gargouri, F.

(2012). A user profile modelling using social

annotations: a survey. In Proceedings of the 21

international conference companion on World Wide

Web, pages 969–976. ACM.

Minamikawa, A. and Yokoyama, H. (2011). Blog tells

what kind of personality you have: egogram

estimation from japanese weblog. In Proceedings of

the ACM 2011 conference on Computer supported

cooperative work, pages 217–220. ACM.

Naghsh, A. M. (2007). SUPPORTING ANNOTATIONS IN

ELECTRONIC PAPER-PROTOTYPING TOOLS. PhD

thesis, Sheffield Hallam University.

Nunes, M. A. S., Cerri, S. A., and Blanc, N. (2008).

Towards user psychological profile. In Proceedings of

the VIII Brazilian Symposium on Human Factors in

Computing Systems, pages 196–203. Sociedade

Brasileira de Computação.

Nunes, M. A. S. N., Cerri, S. A., and Blanc, N. (2008a).

Improving recommendations by using personality

traits in user profiles. In International Conferences on

Knowledge Management and New Media Technology,

pages 92–100.

O’hara, K. and Sellen, A. (1997). A comparison of reading

paper and on-line documents. In Proceedings of the

ACM SIGCHI Conference on Human factors in

computing systems, pages 335–342. ACM.

Omheni, N., Mazhoud, O., Kalboussi, A., and

HadjKacem, A. (2014). Prediction of human

personality traits from annotation activities. In

Proceedings of the 10

International Conference on

Web Information Systems and Technologies, pages

263–269. SciTePress.

Omheni, N., Mazhoud, O., Kalboussi, A.,HadjKacem,

A.(2014). ‘The Annotation: A Track of Reader’s

AutomaticRecognitionofPersonalityfromDigitalAnnotations

279

Personality Traits on Paper’,In: Proceedings of the

ACM Southeast Regional Conference

(ACMSE’14), ACM, article No.10.

Peabody, D. and De Raad, B. (2002). The substantive

nature of psycholexical personality factors: A

comparison across languages. Journal of Personality

and Social Psychology, 83(4):983.

Plimmer, B., Chang, S. H.-H., Doshi, M., Laycock, L., and

Seneviratne, N. (2010). iannotate: exploring multiuser

ink annotation in web browsers. In Proceedings of the

Eleventh Australasian Conference on User Interface-

Volume 106, pages 52–60. Australian Computer

Society, Inc.

Polzehl, T., Moller, S., and Metze, F. (2011). Modeling¨

speaker personality using voice. In INTERSPEECH,

pages 2369–2372.

Prasad, S., Singh, V. K., and Sapre, A. (2010).

Article:handwriting analysis based on segmentation

method for prediction of human personality using

support vector machine. International Journal of

Computer Applications, 8(12):25–29. Published By

Foundation of Computer Science.

Qiu, L., Lin, H., Ramsay, J., and Yang, F. (2012). You are

what you tweet: Personality expression and perception

on twitter. Journal of Research in Personality,

46(6):710–718.

Quercia, D., Kosinski, M., Stillwell, D., and Crowcroft, J.

(2011). Our twitter profiles, our selves: Predicting

personality with twitter. In Privacy, security, risk and

trust (passat), 2011 ieee third international conference

on and 2011 ieee third international conference on

social computing (socialcom), pages 180–185. IEEE.

Rahiman, A., Varghese, D., and Kumar, M. (2013). Habit:

Handwritten analysis based individualistic traits

prediction. International Journal of Image Processing

(IJIP), 7(2):209.

Schiaffino, S. and Amandi, A. (2009). Intelligent user

profiling. In Artificial Intelligence An International

Perspective, pages 193–216. Springer.

Selfhout, M., Burk, W., Branje, S., Denissen, J., Van

Aken, M., and Meeus, W. (2010). Emerging late

adolescent friendship networks and big five

personality traits: A social network approach. Journal

of personality, 78(2):509–538.

Sumner, C., Byers, A., Boochever, R., and Park, G. J.

(2012). Predicting dark triad personality traits from

twitter usage and a linguistic analysis of tweets. In

Machine Learning and Applications (ICMLA), 2012

11th International Conference on, volume 2, pages

386–393. IEEE.

Thomas, L. (2007). Note together: A collaborative

annotation system for the world-wide web. Bachelor’s

thesis, University of Bath.

Tkalcic, M., Kunaver, M., Tasic, J., and Kosir, A. (2009).ˇ

Personality based user similarity measure for a

collaborative recommender system. In Proceedings of

the 5th Workshop on Emotion in Human-Computer

Interaction-Real world challenges, pages 30–37.

Vinciarelli, A. and Mohammadi, G. (2014). A survey of

personality computing.

Wright, W. R. and Chin, D. N. (2014). Personality

profiling from text: Introducing part-of-speech n

grams. In User Modeling, Adaptation, and

Personalization, pages 243–253. Springer.

Wu, W., Chen, L., and He, L. (2013). Using personality to

adjust diversity in recommender systems. In

Proceedings of the 24th ACM Conference on

Hypertext and Social Media, pages 225–229. ACM.

WEBIST2015-11thInternationalConferenceonWebInformationSystemsandTechnologies

280