IBM’s Watson Tone Analyzer is a new service that can analyze text for the attitude and tone that underlie it. The tool could help anyone refine an email, marketing message, presentation or blog post before releasing it into the world said Rama Akkiraju, a distinguished engineer and master inventor for IBM Watson User Technologies.
The goal of Tone Analyzer is to help correct written communication by assessing writing on three primary dimensions. The first is emotional tone, where it focuses on cheerfulness, negative emotions, and anger. Social tone is the second metric it captures by assessing openness, agreeableness and conscientiousness. Finally, writing tone is the third dimension it evaluates in order to provide feedback on how analytical, confident, and tentative the writing is. When it analyzes a passage of text, Tone Analyzer’s results include an explanation of which words contributed to which tone; it also offers alternate word suggestions to refine the text and its tone accordingly.
According to Rama Akkiraju, Distinguished Engineer and Master Inventor at IBM Watson User Technologies, “to read a message and to judge the tone conveyed in the message comes naturally to humans. But, at times, the tone may be overlooked, undesired, or not conveyed well by the author. Can a computer detect the tones conveyed in a message accurately and automatically? Helping humans assess and refine tone in written communication is an interesting challenge in the Artificial Intelligence and Cognitive Sciences fields. At IBM Watson, we are beginning to answer this question.”
Tone Analyzer Scorecard includes Emotional Tone, Social Tone and Writing Style/Tone. For a more elaborate explanation of these ‘tones’ and to test out an analysis of your text, click on the link below for the IBM developer blog.
A study by Northwestern University Center on Media and Human Development of 1,156 adolescents (13-18) studied how teenagers learned about health. It is often assumed that adolescents are easily reached via social media, the researchers found that these efforts do not necessarily true and that often teenagers want to be more private and don’t want others to know what they’re searching about.
Teenagers reported that they primarily looked online for health information for school projects, and then for fitness and diet advice. A third sought information online when a health issue affected them, and a quarter did so to learn more about treating an illness or an injury. One in three teenagers said they changed their behavior because of what they had learned from online sites or apps.
Interestingly, when asked where they got most of their health information, 55 percent of the teenagers cited their parents. School health classes and medical providers also ranked higher than the Internet as preferred sources. Forty-one percent of teenagers searched online for “unhealthy” in formation, including drinking games and how to be anorexic or bulimic, the researchers found. Forty-three percent said they had seen pornography online.
Numerous apps are developed and available for children and many of them are marketed as “educational.” The question is…how do we know which of these thousands of apps will actually help children learn? A report published in Psychological Science in the Public Interest, a journal of the Association for Psychological Science, integrates research from scientific disciplines such as psychological science, linguistics, and neuroscience to assemble an evidence-based guide that parents, educators, and app designers can use to evaluate the quality of present and future “educational” apps.
According to the authors of the report, “The numbers tell the story. Apps are not just ubiquitous, but also big business: Over $10 billion was spent in the App Store in 2013. By 2015, revenue from apps is predicted to triple to $38 billion. Technology is rapidly changing the nature of adults’ day-to-day and even minute-to-minute experiences. We have not begun to understand the impact of the app explosion on our economy and society. The content of the report uses data from decades of research in the Science of Learning to illustrate how the development and evaluation of apps could embrace an evidence-based stance. Importantly, there are a number of theoretical positions on the ways in which children learn.
The researchers conclude that “educational” apps best support learning when they are:
- Active in a way that requires mental effort and not just swiping
- Engaging, not distracting
- Meaningful in the context of a child’s life
- Socially interactive because children learn best with others including parents and peers
“These four principles can help us distinguish apps that masquerade as educational from those most likely to engage children in an educationally meaningful experience,” the authors say. While not all of these principles are necessary, the more an app promotes these types of learning experiences, the greater the educational value of the app will likely be.
According to a new study published in the Proceedings of the National of Sciences (PNAS), computer predictions based on a generic digital footprint (Facebook Likes) are more accurate (r = 0.56) than those made by the participants’ Facebook friends using a personality questionnaire (r = 0.49); (ii) computer models show higher interjudge agreement; and (iii) computer personality judgments have higher external validity when predicting life outcomes such as substance use, political attitudes, and physical health; for some outcomes, they even outperform the self-rated personality score.
The authors point out that “perceiving and judging other people’s personality traits is an essential component of social living. People use personality judgments to make day-to-day decisions and long-term plans in their personal and professional lives, such as whom to befriend, marry, trust, hire, or elect as president. The more accurate the judgment, the better the decision. Previous research has shown that people are fairly good at judging each other’s personalities; for example, even complete strangers can make valid personality judgments after watching a short video presenting a sample of behavior.”
This study compares the accuracy of personality judgment— a ubiquitous and important social-cognitive activity—between computer models and humans. Using several criteria, the researchers demonstrate that computers’ judgments of people’s personalities based on their digital footprints are more accurate and valid than judgments made by their close others or acquaintances (friends,family, spouse, colleagues, etc.). Our findings highlight that people’s personalities can be predicted automatically and without involving human social-cognitive skills.
Smartphone apps are being developed to address mental health issues such as support, reminders, monitoring of mood changes and even stress levels. Tools under development from startups, academic institutions, and research clinics aim to help people manage everything from severe depression to bipolar disorder and schizophrenia. A recent article in Wired magazine points out that “through the discreet and continuous recording of social and physical behavior, these apps can detect changes in mental well-being, deliver micro-interventions when and where needed, and give patients a new awareness of their own illnesses. In the long run, they may even diminish the stigma attached to mental health disorders.”
A Boston-based company Cogito was conducting a clinical trial of an app at the time of the Boston Marathon bombing. Their smartphone application continuously and passively monitors psychological health and well-being using built-in mobile sensors and survey questions. “The bombing offers a unique pre-and post-disaster dataset for understanding the longitudinal trajectories and risk factors for PTSD following trauma. Initial analyses of the data show that survey participation dropped by ~50% during the two weeks immediately after the Boston bombing, perhaps an early indication of the PTSD symptom of withdrawal. Of those that did respond, participants reported an average of ~14% increase in severity of attitudes and behaviors linked with depression and PTSD in the two weeks following the bombing than they did in the two weeks prior.”
A number of apps are being researched to monitor the symptoms of bipolar disorder and schizophrenia. The University of Michigan is working with a company to monitor individuals with rapid-cycling bipolar disorder. Although the possibilities are interesting, many of these ‘medical apps’ are untested and unregulated. Last year the FDA released a document for guidance on medical apps, however it focused mostly on apps that diagnose or turn a smartphone into an EEG. There are no guidelines for mental health apps as yet.
November 20, 2014
A survey by Bank of America states that 47 percent of U.S. consumers admit they wouldn’t last a day without their smartphone. Additionally, of those surveyed who use their phone for mobile banking, 31 percent said they log on at least once a day, while 82 percent access their accounts at least once a week or more. These findings are part of the inaugural Bank of America Trends in Consumer Mobility Report, “a study that explores broad mobile trends and banking behaviors among adult consumers across the country who own a smartphone and have an existing banking relationship. The survey found that mobile connectivity is so critical, the smartphone falls below only the Internet and personal hygiene when ranked by importance to people’s daily lives. Ninety-one percent say their mobile phone is very important, just as important as their car (91 percent) and deodorant (91 percent); and significantly more important than television (76 percent) and coffee (60 percent). The report also found that the youngest millennials ages 18-24 are most likely to view their mobile phones as very important (96 percent) — more so than deodorant (90 percent) and even their toothbrush (93 percent).”
So who own smartphones? The Pew Research Internet Project states that “As of January 2014, 90% of American adults have a cell phone. 58% have a smartphone.”
Here is the breakdown of their study.
Sources: Bank of America Newsroom
According to an article in this month’s JAMA, organized medicine should change the medical profession guidelines on social media use that calls on doctors to separate their personal and professional online personas. The article suggests that “instead of focusing on professional and personal boundaries, physicians should zero in on the boundary between what is appropriate and what is not. When a physician asks, ‘Should I post this on social media?’ the answer does not depend on whether the content is professional or personal but instead depends on whether it is appropriate for a physician in a public space,” the authors said.
The separation of personal and professional personae was a recommendation first made by the American Medical Association in 2010. Other physician organizations followed suit with their own guidelines, each promoting the same personal-professional separation advice. In April, the American College of Physicians and the Federation of State Medical Boards released similar guidelines. According to the JAMA authors, “Separation of identities online is operationally impossible.” They argue that an online search for a social media page would connect both personal and professional pages. “Despite the increasing availability of paid services to monitor and control a person’s Web presence, no current technology exists to overcome fully this particular barrier.”
In addition, the argument highlights that “Professional identity constitutes and is constituted by personal identity, perhaps as one of the many ‘subidentities’ or roles individuals might have, such as a spouse, parent and so on. Separation therefore verges on nonsensical. The argument is also made that separation may be harmful and that depersonalized online interactions could lead patients to believe the doctor is hiding something. They also could lead to a physician’s inability to normalize a difficult situation or express empathy.
Volunteers were recruited to provide information about their feelings and locations. The team created an application for the Android operating system that documented each person’s location and periodically sent the question, “How happy are you?”
Cell phones can efficiently capture information that is generally difficult to record and be “in the moment.” This is important because these notes are likely to be more accurate than feelings jotted down after the fact. Since our society is so mobile and ‘on-the-go,’ this study hopes to capture a range of emotions associated with a geolocation. Over a three-week period, researchers collected information from 270 volunteers in 13 countries who were asked to rate their happiness on a scale of 0 to 5. From the information collected, the researchers created and fine-tuned methods that could lead to a better understanding of how our environments influence emotional well-being. The study was published in the June issue of Demography.
According to the authors, “The mobile phone method could help overcome some of the limitations that come with surveys conducted at people’s homes. Census measurements tie people to specific areas — the census tracts in which they live — that are usually not the only areas that people actually frequent. People spend a significant amount of time outside their census tracks., If we want to get more precise findings of contextual measurements we need to use techniques like this.”
Though many of the volunteers lived in the United States, some were in Australia, Canada, China, France, Germany, Israel, Japan, Norway, South Korea, Spain, Sweden and the United Kingdom. The researchers noted that the team’s focus at this stage was not on generalizable conclusions about the link between environment and happiness, but rather on learning more about the mobile phone’s capabilities for data collection. “I’d be hesitant to try to extend our substantive findings beyond those people who volunteered.” said the lead author Palmer.
For more info and some preliminary results
A study by the Privacy Rights Clearinghouse, a nonprofit advocacy organization in San Francisco, evaluated 43 popular free and paid apps that were made for consumer use. Apps used by health professionals were not part of the study.
Researchers reviewed the mobile application privacy policies and installed and used the apps to see what data were stored on the apps. They also looked at the communication between the apps and the Internet.
The study found that many of the apps sent unencrypted data to advertisers, probably without users’ knowledge. Seventy-two percent of the apps exposed personal information that could include dates of birth, personal location, ZIP codes, medical information, email addresses, first names, friends, interests and weights. Some apps sent information to as many as 10 third parties.
- Data were sent to app developers’ websites and third-party sites for analytic and advertising purposes.
- More than 75% of free apps and 45% of paid apps used behavioral tracking, usually through third parties, according to the study.
Paid apps, which ranged from less than $1.50 to more than $10, posed less risk to users’ privacy. That’s probably because they don’t rely only on advertisers to make money, according to the study. According to AMA News, “Physicians might think twice about advising patients to use some mobile health and fitness apps. A July report indicates that many of those apps compromise patients’ privacy. Just recommending apps may put doctors at risk for violations of the Health Insurance Portability and Accountability Act.”
A study published in Cyberpsychology, Behavior, and Social Networking evaluated 160 Facebook profiles to examine how drinking alcohol is portrayed in both text and images. A second component of the study sought to determine how friends react to these alcohol-related postings.
The analysis generated 2,575 pictures and 92 status updates referring to alcohol use, which represented about 6.50% of the pictures in the total sample and 2.90% of the text updates. These visual and textual references, as well as the peer reactions to these posts, mostly referred to alcohol use in a positive context (72.23% of the pictures and 72.83% of the messages).
The results of this study demonstrate that alcohol-related references are quite common, while negative reactions to these posts are seldom. The authors suggest that further research into the effects of these visual and textual messages is warranted.
A similar study last year in JAMA Pediatrics examined the associations between displayed alcohol use and intoxication/problem drinking (I/PD) references on Facebook and self-reported problem drinking using a clinical scale. The study included undergraduate college students (age range, 18-20 years) at 2 state universities with public Facebook profiles. The profiles were categorized into 1 of 3 categories: Nondisplayers, Alcohol Displayers, and I/PD Displayers.
Results showed that among 307 profiles identified, 224 participants completed an online survey that measured problem drinking using the Alcohol Use Disorders Identification Test (AUDIT) scale. (73% response rate). The average age was 18.8 years; 122 (54%) were female; 152 (68%) were white; and approximately 50% were from each university. Profile owners who displayed I/PD were more likely (odds ratio, 4.4; 95% CI, 2.0-9.4) to score in the problem drinking category of the AUDIT scale, had 64.0% (incidence rate ratio, 1.64; 95% CI, 1.27-11.0) higher AUDIT scores overall, and were more likely to report an alcohol-related injury in the past year (P = .002).
The authors concluded that Facebook displayed references to I/PD were positively associated with AUDIT scores suggesting problem drinking as well as alcohol-related injury. Results suggest that clinical criteria for problem drinking can be applied to Facebook alcohol references.