Towards proposing an intelligent error limiting ... - Semantic Scholar

1

Towards proposing an intelligent error limiting User Interface for rural Indian data entry operators Shrikant Salve, Pradeep Yammiyavar Department of Design Indian Institute of Technology Guwahati, India {s.shrikant,pradeep}@iitg.ac.in

Abstract. This paper intends report an experiment to determine the errors in data entry tasks in the context of rural Indian data entry operations. Rate of errors in data entry is a critical metric for determining output efficiency of data entry operations. Errors in data entry, for example in banking, may result in terrible consequences. It is posited that there is an increase in errors committed by data entry operators when they input data in a language other than their mother tongue (local language). Data entry is a means of lively hood for many rural based service providers in India. This paper studies the errors attributed to local language influences on data entry operators at rural based IT work centers whose manpower is primarily educated in the local language - more often their mother tongue. It proposes an anticipatory warning / correction / error limiting capability to be embedded into the software user interface and to get reflected on the User Interface. Most data entry on computers is done using English language. A pilot study was conducted using an experimental interface designed to test the frequency of errors and difference in 'time required' during English language usage as compared to usage in local language (Marathi or Assamese language in this Indian case). Computer based background recording of data input enabled collection of the speed of input and errors thereof. The results indicate that there indeed is an increase in the errors when the operators use language other than their mother tongue to enter data. Keywords: human computer interaction; data entry; errors; rural; intelligent user-interface.

1

Introduction

It is human tendency to make errors. Any system should be designed to reduce such errors and minimize the consequences when errors happen. Reason (1990) [16] defines errors as “those occasions in which a planned sequence of mental or physical activities fail to achieve its intended outcome, when these failure cannot be attributed to the intervention of some chance agency”. There are two types of errors as slips and mistakes [17]. Slips occur when a person has the knowledge needed to perform a task but because of some reason unintended action happen in completing the task. Mistakes occur when people have incorrect or no knowledge of the task they are completing [19]. In rural India computers are used in many places like banks, railways, bus stands, hospitals, factories, government offices, market places / shops and NGOs (data entry jobs). A task of feeding information in a computer (called as data entry task) has many types of errors called as data entry errors. Simple data entry errors such as typing an incorrect word or number, typing a number twice or skipping a line can give wrong results. In above mentioned context data entry errors can have harsh or serious consequences for rural people who use computers to earn a living. A World Bank [24] report says that in India, almost 72% of the population stays in villages. Also 22 regional and 2 national languages that is Hindi and English are spoken but English is not an official language [11], [21]. About 92.39% schools in rural areas teach in the medium of regional languages (mother tongue) [12]. The language used for data entry in most of above mentioned places in rural India is observed to be English, expect in few NGOs and government offices. In the rural part of Indian the main problem is illiteracy because of poverty [10]. However most information systems are in the English language. Development cost of applications with community partners that meet their local language learning needs, is beyond the budgets of community development projects. In such a scenario the reliability and quality of rural based data entry services may also become questionable. Human Computer Interaction (HCI) is emerging field of study that has begun to explore methods to improve a computer user interface for rural people through multilingual support. Data entry mechanism becomes pervasive in the area of HCI. Data entry technologies are designed and evaluated through empirical evaluation with intervention of user. Designing of effective data entry mechanism are mainly focused on two parameters as speed and accuracy or errors [9]. Our study aims at finding efficiency (speed and error rate) of data entry by rural based users. The intention is to work towards proposing an intelligent error prompting or correcting software interface that can anticipate known errors based on contextual analysis and warn or prompt the operators or their supervisors while engaged in data entry.

Volume 13, No. 4

Australian Journal of Intelligent Information Processing Systems

2

2

State of the Art

Chand et. al. [4] have proposed Jadoo- A paper user interface for people living in rural India. It is a prototype system used by computer literate to create and distribute paper user interface which can be used by computer illiterate to access online information. He stated that illiteracy, user of a non-native language and fear of technology are big hurdles for rural users in India. Patel et. al. [14], [15] have developed voice-based community forum (named Avaaj Otalo) interface for rural people in India. This application was developed in Gujarati language which allows farmers to receive timely and relevant agriculture information over the phone. They stated importance for spoken language research in rural areas of the developing world. Chand [5] has discussed issues involved in designing the interface for computer driven kiosks used in rural areas of India. He stated that the content of such interface for rural users should be developed using images, multilingual text and videos in order to support illiterate and multilingual users. He analyzed user-kiosk interface based on factors like: motivation, visual interface, mouse based interaction, navigation, media (videos, animations, text and images). His study raises the importance of multilingual text and video contents while developing interface for rural people. Gore et. al. [10] have proposed GappaGoshtiTM (mobile based platform) – a social collaborative system to exchange information amongst rural people. They stated that, this social collaborative system could be well adapted by rural people if provided with local language along with valuable information. Barchard et. al. [2] have projected the study on impact of human data entry errors on statistical results and calculations. They used 195 undergraduate students to participate in experiment by assigning three data entry methods- double entry, visual checking and single entry to them. The participants entered 30 data sheets, each containing six type of data. Their results show that in double entry significantly fewer errors than both visual checking and single entry. Oladimeji et. al. [13] have proposed the study of number entry interface found on medical devices. They reported an experiment that investigates the effect of interface design on error detection in number entry tasks using two number entry interfaces, one serial interface with 12 key numeric keypad and another incremental interface that use a knob or a pair of keys to increase or decrease numbers. 22 participants aged 18-55 years took part in the experiment. A computer was used with an integrated eye-tracker to present the instructions and number entry interfaces. Each participant used both number entry interfaces (independent variable). The dependent variables were the number of undetected errors, number of corrected errors, total eye fixation time and task completion time. The participant was required to enter 100 numbers using both interfaces according to the instruction shown on the right half of the screen. They identified six categories of number entry errors (skipped, transposition, wrong digit, missing decimal, missing digit and other). They have suggested giving priority to research number entry styles and their relation to error rate, behavior and performance in the context of safety critical number entry systems. However their study restricted to medical number entry systems. Published literature in which studies on Indian local language figure are very few in number. One of them by Ghosh et. al. [9] reported on the cost of error correction during Bengali Text transcription. Ghosh et. al. have proposed algorithms on the error correction quantification problem for Indian language (Bengali) transcribed text typed by any single stroke or tap text entry tool. They identify the unequal character (error) positions in both transcribed and presented text both by using longest common subsequence algorithm. They proposed two algorithms first to identify whether error in simple character or in complex character. Second to calculate the minimum number of operations to renew transcribed to presented text depending upon error positions and type (error in simple or complex character). They also define correction cost per error metric to calculate average correction cost for an erroneous transcribed text. This is the only study based on text entry errors for Indian language. This paper also highlights the complexity of Indian languages because of complex typography styles. Figure 1 is an example of the complex typography of Indian language Hindi in Devanagari script. Combining multiple characters like vattu (5) a below base form of a consonant as in figure 1 and matra (4) makes it difficult to measure the errors in transcribed texts.

Fig. 1. Example in Devanagari script [8]

Australian Journal of Intelligent Information Processing Systems

Volume 13, No. 4

3

Singh et. al. [20] have proposed study on the numeric paper forms used by the NGOs for data collection in rural India. They have investigated NGO’s form filling requirements which were used to interact with rural people. They proposed the numeric input method for different NGO’s form filling requirements which is easy to use for rural people and also machine readable. This paper's context is the data entry jobs provided for rural people by various NGOs in India. Chen et. al. [6] have proposed USHER- it is based on probabilistic model over the questions of the data entry form, which learns from previous form submission. During data entry USHER dynamically adapt the form to the values being entered by providing real time interface feedback. They evaluated USHER using real world patient data set at an HIV/AIDS program in Tanzania. Their work gives motivation towards the design of intelligent data entry interface for rural Indian users. The literature says that in rural India people have minimum access and familiarity with computers because of illiteracy and spoken language problem [4], [5], [9], [10], [14]. This has created a digital divide between the information haves and have-nots. In this paper, our study focuses on highlighting regional language induced problems in rural India. We study the performance of users carrying out a numerical entry and text entry task using two different languages namely English and local language. We report our findings on the relative accuracies of both languages and classification of the types of errors made by rural Indian users who were subjects of our experiment. Participants with Indian rural background were given designed tasks to perform. There were two types of tasks designed as part of the experimental interface instrument. The first involved use of calculator interface for numerical data entry and second was the use of Microsoft Word and BarahaPad for word / text data entry. Two local Indian languages (Marathi & Assamese) were chosen for experimentation due to the researches being in there geographical proximity.

3

Research Hypothesis

The state of the art depicts attention towards data entry errors are vital for evaluating efficiency of rural users while interacting with computers. We are investigating the influence and extent of contribution of local language in triggering these errors. We proposed the following hypothesis: H1- The rural users make more errors in English numerical data entry than while using local language (Marathi and Assamese) for numerical data entry. H2- Rural users require more time for typing English numerical data than if they do it using local language (Marathi and Assamese) methodology. H3- Rural users make more errors in text entry using English language as compared to local language (Assamese). H4- Rural users require more time in text entry in English language as compared to local language (Assamese). The usability experiment research design set up to test the above hypothesis is given below.

4

Research Design

We categorized data entry operation as Numerical (1, 2, 3…) entry and Text entry. Separate experiments were designed for each category of operation and named as experiment I and experiment II. Table 1 shows the research design adopted for the experiment along with corresponding hypothesis under test. Table 1. Research Design Experiment I H1- The rural users make more errors in English numerical data entry than local language (Marathi & Assamese) numerical data entry H2- The rural users require more time in typing English numerical during data entry than if they do it using local language (Marathi & Assamese) methodology Participants Sample size : 48 subjects Instruments CALCI software interface (specifically conceived for this experiment) used Independent variables- numerical entry lanResearch guage (three types): English language interface model and two local languages interface. Dependent variables- task completion time and errors Hypothesis

Volume 13, No. 4

Experiment II H3- Rural users make more errors in text entry in English language compare to local language H4- Rural users require more time in text entry in English language as compared to local language (Assamese)

Sample size : 20 subjects Microsoft Word and BarahaPad (typing software similar to Notepad.) Independent variable- text entry language (two types): English language interface and Assamese languages interface Dependent variables- task completion time and errors

Australian Journal of Intelligent Information Processing Systems

4

Experiment I: A software instrument named here as CALCI was specifically conceived and designed using Visual Basic .Net (version 2010) for numerical data entry tasks using keyboard and mouse input device. The CALCI does calculation using three different languages English, Marathi and Assamese. Experiment II: Microsoft Word software for English language text entry and BarahaPad [1] for rural Indian local language - Assamese for text entry. 4.1

Participants

Total Sample size of 68 participants (male and female) were approached randomly with a request to participate. The subjects belonged to age group 18 to 30 years. They worked in shops, vegetable market and as security guards in the campuses of Indian Institute of Technology Guwahati (IITG). Of the 64 subjects, 24 belonged to rural villages in the western Indian state of Maharashtra which was physically visited by the researchers for collecting data. Figure 2 depicts pictures of the participants performing the experiment. Pictures of participants are used with their consent.

Fig. 2. Users performing the data entry operation

Demographics Participants (63 Males + 5 Females) of age group 18 to 30 years had educational qualification varying between 10th to th 12 standard of high school level (that is non-graduate) and used computers or laptops at least one hour in a week as part of their jobs or for personal communication use. Local Language and English Baseline All the participants had completed schooling in their mother tongue language (either Assamese or Marathi). Their proficiency in English was ascertained before the test. This was done by giving a sentence in English and asking subjects to translate it into their local language. Following are the errors and problems observed in translation of given English sentence into Assamese language by rural users with their percentages expressing their proficiency in English. This proficiency will be used later on to compare with the test results.    

Spelling errors (35%) Grammatical errors (55%) Difficulty in phrasing complete sentences (20%) Difficulty in locating the keys for typing complex words (combining multiple characters together) in case of Assamese language (75%).

Technology Baseline Among the 68 participants, only 20 of them have access to computer / laptop at their homes. All the participants uses computer / laptop at least one hour in a week.

Australian Journal of Intelligent Information Processing Systems

Volume 13, No. 4

5

4.2

Instruments used

Experiment I: CALCI: A software interface was designed especially for this experiment using Visual Basic language on Visual Studio .Net platform. It aided numerical data entry using both key board and mouse as input device. It performs and displays calculations using three different languages English, Marathi and Assamese through the User Interface. This instrument named as CALCI does arithmetic operations such as addition, subtraction, multiplication and division on numbers with and without decimal point. Figure 3 depicts three screen shots of English, Assamese and Marathi language calculator interface respectively.

Fig. 3. Screen shot of software interface designed, which does calculations in three languages - English, Assamese & Marathi

Experiment II: Microsoft Word for English language text entry was adopted and for Assamese language text entry- a locally developed Notepad equivalent software named as BarahaPad [1] was utilized. Figure 4 shows the keyboard design which can be used to type in both English as well as Assamese language. The keyboard of Baraha is depicted in figure 4.

Fig. 4. Keyboard for Assamese and English language text entry

4.3

Research Model

Both experiments were designs using ‘within subject repeated measures'. Experiment I: The participant used CALCI interface six times to complete six tasks for calculation in English and local languages. The participants perform same calculation in English and local languages using two input devices namely mouse and keyboard. The numerical entry was the independent variable and it had three types: English language interface and two local languages interface. The dependent variables were the task completion time and errors made in numerical data entry. Experiment II: The text entry language was the independent variable and it had two types: English language interface (that is Microsoft Word) and Assamese languages interface (BarahaPad). The dependent variables were the task completion time and errors made in text entry. 4.4

Procedure

All participants were tested independently. Experiment I: The CALCI interface was used for each participant and they were briefed about the stages and purpose of the experiment before starting. The experiment was divided in six parts (tasks) as given in table 2. In table 2: Task 1 consists of local language numerical data entry by using input device mouse without time limit, task 2 consists of local language numerical data entry by using mouse within limited time (that is one minute) and so on. Each participant has to perform all tasks, but the sequence/order of the tasks may be different.

Volume 13, No. 4

Australian Journal of Intelligent Information Processing Systems

6

Table 2. Task design for experiment I. Task No. Task 1(T1) Task 2(T2) Task 3(T3) Task 4(T4) Task 5(T5) Task 6(T6)

Language Local Local Local English English English

Input Device Mouse Mouse Keyboard Mouse Mouse Keyboard

Time Allotted Without lime limit Within one minute Within one minute Without time limit Within one minute Within one minute

Table 3. Equal distribution of samples among each task Sample Distribution 1-8 9-16 17-24 25-32 33-40 41-48

Local Language T1 T2 T3 1 3 5 5 1 3 3 5 1 2 4 6 6 2 4 4 6 2

English Language T4 T5 T6 2 4 6 6 2 4 4 6 2 1 3 5 5 1 3 3 5 1

Table 3 shows how samples distribution was done among each task and sequence of tasks to perform. The first row of table 3 consist of 1-8 samples were performed task1 to task 6 in sequence/order first task 1, second task 4, third task 2, fourth task 5, fifth task 3 and sixth task 6 (i.e. 1-3-5-2-4-6, see table 3’s first row) and likewise. Prior to each stage of the experiment, the participants were given orientation session where they could enter 2/3 simple calculations and get familiar with the interface. When the participants were comfortable with how the interface worked, they were allowed to proceed to the experiment. The participants were required to enter given mathematical calculations having three different difficulty levels (like very easy, easy and hard) using two interfaces (English and Marathi or Assamese) in the defined order in table 3. The participants were provided the experiment sheet including mathematical calculations in English and local language they speak. Experiment II: Each participant was used BarahaPad and Microsoft Word for Assamese and English language text entry respectively. Table 4 depicts the task design for text entry experiment. Table 4. Task design for experiment II. Task No. Task name Experiment tool Task 1 Type given Assamese language sentence BarahaPad Task 2 Type given English language sentence Microsoft Word

To get familiar with the interfaces the participants were given an orientation session where they could enter given sentences in both languages (English & their local language) so that the participants were comfortable with the experimental instrument interface. Assamese language contains thirty one phonemes, eight vowel and other twenty three consonant phonemes [18]. The phonetic keyboard was designed by sticking the Assamese language phonemes on keys of regular English language keyboard (see figure 4). Writing Assamese words using phonetic keyboard is as easy as writing in English. The participants were instructed to perform the tasks (data entry) as quickly and as accurately as possible. The computer based background recording of each participant interaction with interfaces have taken to collection of speed of entry and errors.

5

Results and Discussion The table 5 illustrates the statistical analysis of results obtained from both experiments. Table 5. Statistical anaylasis of results Hypothesis t- value mean (paired t-test) English Local H1 -3.45 0.52 0.35 H2 -2.69 159.9 152.5 H3 3.56 5.75 4.95 H4 -1.90 32.20 32.85

Australian Journal of Intelligent Information Processing Systems

sd English Local 0.69 0.49 80.67 80.13 2.57 2.46 4.79 4.88 Volume 13, No. 4

7

5.1

Errors

Experiment I: Analysis of the total numerical data entry errors for both interfaces (English and local language) using paired t-test indicates that the mean errors for the local languages numerical data entry (mean=0.35, sd=0.49) was significantly lower than that of the English language numerical data entry (mean=0.52, sd=0.69), t(144) = -3.45, p < 0.001. Below, we report the different types of error that occurred in our experiment. Wiseman et. al. [23] and Oladimeji et. al. [13] have proposed a classification of numerical entry errors. They reported the occurrence of certain error types between the two numerical entry interface styles. The classification and frequency of each group of numerical entry errors is shown in figure 5. 40

36

30

English Local 22

20

15 14 9

10

13

8

6

2

1

0 Decimal pointMissing digit Wrong digit Double entry Reverse digit Fig. 5. Classification of numerical entry errors

There are common user errors observed in numerical entry interfaces experiment. 1. Decimal point: This error occurs when a decimal point is absent or misplaced from the transcribed number but is present or appropriate in the instruction. 2. Missing digit: This refers to occurrences of errors where one digit, one digit before and after decimal from the intended value missing from the transcribed value. 3. Wrong digit: Wrong digit errors occur when one of the digits in the written value is incorrect. 4. Double entry: This type of errors occurs when double or repeated entry of numbers is found. 5. Reverse digit: Reverse digit or transposition happen when the user switches two adjacent digits in a number. Experiment II: Error rate- We used Levenshtein minimum string distance statistic for measuring error rates in text entry [3], [22]. Statistical analysis of the total text entry errors for both language entry (English and Assamese) using paired t-test indicates that the mean errors for the English language text entry (mean=5.75, sd=2.57) was significantly higher than that of the Assamese language text entry (mean=4.95, sd=2.46), t(20) = 3.56, p < 0.002. Types of errors observe in text entry by rural users, 1. Spelling, Incorrect/ Missing Word: This error occurs when user forget character within words or whole word in transcribed text. 2. Double character: In this type user try to create an unwarranted duplicate character after a target character. 3. Unrelated: This error means creating unrelated characters in relation to the presented text. 4. Related: This error means deleting characters that are related to the presented text. 5. Case: This refers to entering a target character in the wrong case (that is creating a uppercase letter when it is supposed to be in uppercase, or vice versa). 6. Layer switching: User needlessly switching between the upper / lower case layer of the keyboard (e.g. by pressing SHIFT / CAP lock). Below figure 6 depicts the taxonomy of text entry errors by rural users,

50 40

44

English Assamese

33

30 20

9 10

10

11 9

9

22 21

20 19

Case

Layer switching

7

0 Spelling

Double Unrelated Related character

Fig. 6. Taxonomy of text entry errors

Volume 13, No. 4

Australian Journal of Intelligent Information Processing Systems

8

5.2

Task completion time

It is the time taken by participants to complete a task. Experiment I: Task completion time of Task 1 for English and Task 4 for local language was measured. The task completion time for English language numerical entry (mean=159.9, sd=80.67) was significantly slower than the task completion time for local language numerical data entry (mean=152.5, sd=80.13), t(48) = -2.69, p < 0.01. Experiment II: Task completion time of Task 1 for Assamese and Task 2 for English language was measured. The task completion time for Assamese language text entry (mean=32.85, sd=4.88) was not significant compared to the task completion time for English local language text entry (mean=32.20, sd=4.79), t(10) = -1.90, p > 0.073. 5.3

Other Observations and Findings

Experiment I: The number of errors per participants within limited time by input device mouse (mean=0.53, sd=0.65) was significantly more than without time limitation (mean=0.28, sd=0.54), t(96) = -3.14, p < 0.002. Another analysis of the total number data entry errors for both local languages number data entry (Marathi and Assamese) using independent t-test shows that total number of errors in Assamese language numerical entry (mean=0.46, sd=0.56) were more in comparison with Marathi language numerical entry (mean=0.25, sd=0.44), t(72) = -2.50, p < 0.01. 5.4

Discussion

Experiment I: The results show a considerably higher number of errors on the English numerical data entry task in comparison to the local language (Marathi & Assamese) numerical data entry task. It was also observed that rural users were slower (37.5 seconds) during data entry using English as compared to while using local language. Large difference are noticed in ‘Wrong digit’ errors group which may be due to higher familiarity and understanding of local language numerical by rural users. For both, English and local languages numerical entry task, number of errors within limited time entry was significantly more than without the constraint of time limit. This was probably because the participants try to key in the entries faster within the given limited time. This results in more errors in comparison to when there is no time limit prescribed. The results also show that users of Assamese language make more errors compared to users in Marathi. The statistical analysis results indicate that significant differences in the error rates and slight different in speed of entry for the two experimental conditions of numerical entry. This upholds both hypothesis (H1 and H2). Experiment II: The results confirm a noticeably higher number of errors on the English text entry task in comparison to the local language (Assamese) text entry task. It was also observed that rural users were slower during text entry using Assamese language as compared to while using English language. It was noticed that during typing rural user reads the presented text and then types it using the given interface. While reading the presented Assamese word from sentence, he try to memories 2-3 words and then type it on the screen. But in case of English language words (specially long and difficult words) the rural user is unable to remember the full spelling while transcribing and makes more errors. The structure of Indian languages are different from English containing simple, complex and matra characters. The complex character is made by combining multiple characters together and error in one single character may be required multiple edit operation to fix it. In such case fixing of a single error requires additional edit operation to non-erroneous characters. So for Indian language (Assamese) the number of edit primitives required to transform transcribed text from presented text is more- which indeed requires more time as compared to English. The results of statistical analysis indicate that significant differences in the error rates exists which upholds hypothesis (H3) (Table 1 & Table 5). However we found no significant difference in speed of entry for the two experimental conditions of text entry (hypothesis (H4)) thereby resulting in failure to reject the null hypothesis.

Conclusion and Future Work Experiment I- There are significant differences in the error rates and slight different in speed of entry for the two experimental conditions of numerical entry. The rural users made more errors and required more time in English numerical data entry than data entry using local language (Marathi and Assamese language). Experiment II- There is significant difference in the error rates for the two experimental conditions of text entry. The rural users made more errors in English text entry than local language (Assamese).

Australian Journal of Intelligent Information Processing Systems

Volume 13, No. 4

9

The result suggests that for a designer involved in designing interfaces or navigation for predominantly rural users more comfortable with local language, influence of local language needs to be taken into account while determining the information architecture in an application.

6

Intelligent error limiting user interface for data entry : a proposal

An intelligent prompting cum self-correcting graphic user interface with the schema in figure 7 is proposed based on the findings of the experiments presented in this paper.

Fig. 7. Intelligent error limiting user interface (IELUI)

In rural India data entry is done at many Rural BPO’s (like RuralShores, Sourse2Rural etc.) using paper forms for data entry via computers. We have adopted the technique of Chen [6], [7] known as USHER which provides data quality with data entry forms. We named it as 'Intelligent Error Limiting User Interface (IELUI)'. It is probabilistic model based on various fields for data entry form logging. IELUI uses standard machine learning techniques (structured learning) to induce probabilistic model from previous form entries. The probabilistic model of the data represented as a Bayesian network over form fields. This network captures relationships between a form’s field’s elements in a stochastic manner which then allows us to generate predictions and error probabilities for the form. For example, given input values for some elements of the fields of a particular form, the model can infer probability distributions over values of that instance and prompts the most likely items (type ahead suggestions and default values). IELUI will also provide feedback in terms of message and voice in their local language which then will be used to display errors or prompt messages to the operator whenever an error is committed.

References 1. BarahaPad- http://baraha.com 2. Barchard, K.A., Pace, L.A.: Preventing Human Error: The Impact of Data Entry Methods on Data Accuracy and Statistical Results. Computers in Human Behavior, vol. 27, issue 5, pp. 1834-1839 (2011) 3. Calculate Levenshtein Distance, http://www.unit-conversion.info/texttools/levenshtein-distance, Retrieved on date 10 October 2013 4. Chand, A., Dey, A.K.: Jadoo: a Paper User Interface for Users Unfamiliar with Computers. Proc. CHI 2006, pp. 1625-1630 (2006) 5. Chand, A.: Designing for the Indian Rural Population: Interaction Design Challenges. Proc. Development by Design Conference, (2002) 6. Chen, K., Chen, H., Conway, N., Hellerstein, J.M.; Parikh, T.S.: Usher: Improving Data Quality with Dynamic Forms. Knowledge and Data Engineering, IEEE Transactions on. vol.23, no.8, pp.1138-1153 (2011) 7. Chen, K., Parikh, T., Hellerstein, J.M.: Designing Adaptive Feedback for Improving Data Entry Accuracy. Proc. ACM Symp. User Interface Software and Technology (2010) 8. Developing OpenType Fonts for Malayalam Script. http://www.microsoft.com/typography/OpenTypeDev/malayalam/intro.htm. Retrieved on 18 October 2013 9. Ghosh, S., Samanta, D., Sarma, M.: Cost of Error Correction Quantification with Bengali Text Transcription. Intelligent Human Computer Interaction (IHCI), 4th International Conference, pp.1-6, IEEE (2012) 10. Gore, K., Lobo, S., Doke, P.: GappaGoshtiTM: Digital Inclusion for Rural Mass. Communication Systems and Networks (COMSNETS), pp.1-6, IEEE (2012)

Volume 13, No. 4

Australian Journal of Intelligent Information Processing Systems

10

11. Kam, M., Kumar, A., Jain, S., Mathur A., Canny, J.: Improving Literacy in Rural India: Cellphone Games in an After - School Program. Information and Communication Technologies and Development (ICTD’09), pp. 139-149. IEEE Press, Piscataway, NJ, USA (2009) 12. Meganathan, R.: English Language Education in Rural Schools of India: the Situation, the Policy and the Curriculum. BBC World Service, October, 2009, http://www.teachingenglish.org.uk/blogs/rama-meganathan/english-language-education-rural-schools-india-situation-policy-curriculum 13. Oladimeji, P., Thimbleby, H., Cox, A.: Number Entry Interface and their Effects of Error Detection. INTERACT (2011) 14. Patel, N., Agarwal, S., Rajput, N., Nanavati, A., Dave, P., Parikh, T.S.: Experiences Designing a Voice Interaction for Rural India. Spoken Language Technology Workshop, SLT 2008. pp. 21-24. IEEE (2008) 15. Patel, N., Chittamuru, D., Jain, A., Dave, P., Parikh, T.S.: Avaaj Otalo- A Field Study of an Interactive Voice Forum for Small Farmers in Rural India. CHI 2010. ACM. Atlanta, GA, USA (2010) 16. Reason, J.: Human Error. Cambridge Press (1990) 17. Reason, J.: The Human Contribution. Farnham, UK: Ashgate (2008) 18. Sarma, M., Sarma, K.K.: Segmentation of Assamese Phonemes using SOM. In Emerging Trends and Applications in Computer Science (NCETACS), 2012 3rd National Conference on, pp. 121-125. IEEE. (2012) 19. Sauro, J.: Measuring Errors in the User Experience. Measuring Usability (2012) 20. Singh, G., Findlater, L., Toyama, K., Helmer, S., Gandhi R., Balakrishnan, R.: Numeric Paper Forms for NGOs. In Proc. ICTD (2009) 21. Smith, A., Joshi, A., Liu, Z., Bannon, L., Gulliksen, J., Christina Li.: Institutionalizing HCI in Asia. In Proceedings of the 11th IFIP TC 13 International Conference on Human-computer Interaction - Volume Part II (INTERACT'07) (2007) 22. Soukoreff, R.W., MacKenzie, I.S.: Measuring Errors in Text Entry Tasks: An Application of the Levenshtein String Distance Statistic, Extended Abstracts of CHI 2001. pp. 319-320. ACM, New York (2001) 23. Wiseman, S., Cairns P., Cox, A.: A Taxonomy of Number Entry Error. HCI 2011: The 25th BCS Conference on Human-computer Interaction (2011) 24. World Bank, Rural India Population, http://go.worldbank.org/8EFXZBL3Y0, Retrieved on 26 August 2013

Australian Journal of Intelligent Information Processing Systems

Volume 13, No. 4